 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. Florid Reactive Periostitis Ossificans of the Proximal Aspect of the Tibia: A Lesion That Must Be Distinguished from Osteosarcoma. A Case Report* , LOS ANGELES, CALIFORNIAInvestigation performed at the Orthopaedic Oncology Service, Orthopaedic Hospital, Los Angeles
Florid reactive periostitis ossificans is a benign lesion that was first described, to the best of our knowledge, by Spjut and Dorfman in 198113. It has been reported exclusively in the tubular bones of the hands and feet1-8,10-14. Typically, this lesion affects adolescents and young adults and presents as a swollen, painful, and erythematous finger or toe. Spjut and Dorfman reported that the lesion appears benign radiographically but could be mistaken for a variety of malignant lesions, especially osteosarcoma, on histological examination13. Magnetic resonance imaging scans often demonstrate massive signal changes within the soft tissue that are consistent with infection, trauma, or a tumor of the soft tissue or bone. We report the case of a five-year-old girl who was followed clinically and radiographically after the diagnosis of florid reactive periostitis ossificans involving the proximal aspect of the tibia. To our knowledge, we are the first to describe the occurrence of this lesion in a long bone.
A five-year-old girl was seen because of a two-month history of progressive pain in the left knee that was increasing in frequency and duration. The pain was aggravated by walking, and the patient was unable to bear full weight on the extremity. There was no history of fever, chills, night sweats, or injury. Physical examination revealed fullness and tenderness in a six by three-centimeter area about the proximal-medial aspect of the posterior part of the left tibia. Measurement of the circumferences of the thighs demonstrated 1.5 centimeters of atrophy of the quadriceps muscle on the left. The lower extremities were of equal length. The circumference of the proximal aspect of the calf was bilaterally symmetrical. Neurovascular function was intact distal to the lesion, and there was no evidence of adenopathy proximal to the lesion. Laboratory studies revealed a normal white blood-cell count of 9.6 x 1010 per liter, with 61 percent neutrophils and 32 percent lymphocytes, and an elevated erythrocyte sedimentation rate that fluctuated between thirty-two and seventy-five millimeters per hour. The rheumatological workup was positive only for HLA B-27. The rest of the laboratory findings, including serum levels of calcium (10.6 milligrams per deciliter [2.64 millimoles per liter]), phosphorus (4.7 milligrams per deciliter [1.52 millimoles per liter]), and alkaline phosphatase (149 international units per liter), were within normal limits. Plain radiographs showed a 2.0 by 1.5-centimeter blastic lesion about the proximal-medial aspect of the tibia, with central irregularity, mild expansion of the posterior aspect of the proximal tibial metaphysis, and a large associated soft-tissue mass (Fig. 1). T2-weighted magnetic resonance imaging scans revealed heterogeneous signal intensity within the soft tissues surrounding the proximal aspect of the tibia, with accompanying changes in the marrow (Figs. 2-A and 2-B). Triple-phase bone-scanning showed increased uptake in the proximal aspect of the left tibia only.
Two months before the patient was seen by us, a core biopsy of the proximal aspect of the tibia had been performed at another institution and was found to be nondiagnostic. Specimens of the lesion were obtained for aerobic, anaerobic, mycobacterial, and fungal culture and were plated on blood, Thayer-Martin, Löwenstein-Jensen, and Sabouraud dextrose agars, respectively. All cultures were negative. Computerized tomography showed a sclerotic lesion measuring 2.5 centimeters in diameter, with posteromedial expansion of the proximal aspect of the tibia and an associated soft-tissue mass. When the patient was seen by us, samples of bone were obtained with a curet from the center and periphery of the proximal aspect of the tibia and biopsy specimens were obtained from the surrounding soft tissues. Specimens of the lesion were once again taken for aerobic, anaerobic, mycobacterial, and fungal culture and were plated on blood, chocolate, Löwenstein-Jensen, and Sabouraud dextrose agars, respectively. All cultures were negative. Grossly, the soft-tissue specimens were gray-tan, yellow-tan, or pink-tan. Histological analysis revealed evidence of woven bone, foci of cartilage production, and areas of fairly intense osteoid production with slight cellular atypia. In some areas, the abundant osteoid production resembled that seen in association with osteoid osteoma, osteoblastoma, fracture callus, and osteosarcoma. However, no definite anaplasia was observed. Under high-power magnification, stromal cells in this region demonstrated mild nuclear atypia without osteoblastic rimming. The nuclear chromatin stained distinct and crisp (Figs. 3-A and 3-B). A richly myxoid to myxochondroid background was seen adjacent to the primitive osteoid (Figs. 4-A and 4-B).
The diagnosis of florid reactive periostitis ossificans was made on the basis of these findings. The patient initially was managed with intravenous administration of cefazolin (650 milligrams per day), but this treatment was discontinued after two days because the cultures remained negative. No additional treatment was given. At the time of the thirteen-month follow-up, magnetic resonance images showed normal signal intensity within the soft tissues and marrow. By approximately fifteen months, the patient had resumed full activities and had no residual swelling or pain. Plain radiographs that were made at that time revealed that the soft-tissue mass had resolved and that the proximal aspect of the tibia appeared more normal (Fig. 5).
The recommended treatment of florid reactive periostitis ossificans generally includes local excision and close follow-up. Spjut and Dorfman described twelve patients who had a lesion involving the tubular bones of the hands and feet13. In each instance, the histological appearance of the lesion resulted in a high suspicion of malignancy; however, ten of the lesions were treated with local excision, one was treated with amputation of a ray, and one was treated with biopsy followed by observation. All twelve patients had an excellent result with no recurrence13. The clinical findings in our patient included severe pain, an inability to bear full weight on the extremity, and generalized swelling about the proximal aspect of the tibia. The radiographic and scintigraphic findings, in conjunction with the signal abnormalities within the soft tissues and marrow as seen on magnetic resonance imaging scans, were consistent with those associated with a variety of infectious, traumatic, benign extracapsular destructive, or malignant bone and soft-tissue lesions. Each of these lesions can cause changes in relative fluid content, which may result in signal abnormalities on magnetic resonance imaging scans. The histological finding of abundant osteoid production was consistent with a primary bone-producing neoplasm. We recently managed a patient with florid reactive periostitis ossificans of the phalanx who had clinical, radiographic, and histological findings that were nearly identical to those described in the present report. Although we are not aware of any previous reports of florid reactive periostitis ossificans of a long bone, the findings in the patient described in the current report were thought to be consistent with such a diagnosis. The radiographic appearance of florid reactive periostitis ossificans is generally that of a benign lesion with evidence of soft-tissue swelling, which is sometimes associated with soft-tissue calcification. Many other local and systemic factors may contribute to reactive periostitis. A periosteal reaction may be present but the osseous cortex is usually intact, without evidence of osseous destruction. However, we are aware of five patients, in addition to the one described in the present report, who had cortical erosion in association with a periosteal reaction, giving the lesion the appearance of a rapidly growing and potentially malignant tumor3-5,7,13. All five patients were successfully managed with excision through the reactive margin of the lesion. Cortical erosion is not always a reliable indicator of severity or malignancy. Bone scintigraphy has been reported to show increased uptake in local lamellar bone8, but this has not been our experience in the evaluation of florid reactive periostitis ossificans. Nora disease9, a bizarre, periosteal osteochondroma-like proliferation, is also most commonly found in the hands and feet and may demonstrate histological atypia similar to that seen in association with florid reactive periostitis ossificans. However, Nora disease is a heavily calcified juxtacortical lesion that demonstrates a unique histological finding of blue-staining woven bone. In our patient, diffuse involvement of the intramedullary and extramedullary portions of the proximal aspect of the tibia was identified. Stress or fatigue fracture also may have radiographic and histological features that are similar to those of florid reactive periostitis or neoplasm. The radiographic findings associated with a stress fracture may include a hair-thin crack, periostitis, and late callus formation. In addition, magnetic resonance images often show marrow edema surrounding a fracture line (indicated by a dark linear band between the normal and edematous regions) as well as high signal intensity within the adjacent soft tissues. The radiographic findings in our patient did not include a fracture line, late callus formation, or a dark linear band on magnetic resonance imaging scans. Histologically, fracture callus may be misinterpreted as florid reactive periostitis or osteosarcoma if a biopsy is performed within the first two to four weeks after the fracture. By two months, however, the osteoid and woven bone that are found in fracture callus would be expected to have matured into lamellar bone with prominent foci of osteoblastic rimming and conversion of stroma back into fat. In our patient, the histological findings at two months revealed abundant myxoid stroma, no lamellar bone, and no osteoblastic rimming. Consequently, stress fracture was ruled out both radiographically and histologically. The histological appearance of florid reactive periostitis ossificans is characterized by the predominance of active, large, spindle-shaped fibrous cells with prominent nuclei. Mitoses may be frequent but not abnormal. Some areas may include both osteoid and myxoid elements. Multinucleated giant cells and inflammatory cells are also seen frequently13. These histological features may be confused with those of infection and a variety of malignant processes, especially high-grade osteosarcoma. A diagnosis of osteosarcoma may be considered because of the infiltrative nature of the lesion, the plump appearance of the fibroblast-like cells, and the presence of osteoid. The correct diagnosis can be made if particular attention is paid to the lack of true anaplasia of the spindle-cell elements and the absence of atypical mitoses. In the case of our patient, the hypocellular myxoid component seen on histological studies helped to establish the diagnosis of florid reactive periostitis ossificans. The clinical, radiographic, and histological features of the lesion in our patient supported a diagnosis of florid reactive periostitis ossificans. In general, the surgeon's objective should be to obtain representative histopathological samples from the central and peripheral parts of the lesion and to label them accordingly. This practice is particularly important for differentiating myositis ossificans from soft-tissue osteosarcoma. Adequate sampling is determined by intraoperative consultation with the pathologist, who determines whether the findings of histological analysis of frozen sections represent a preliminary tissue diagnosis. Local excision appears to be the preferred treatment of florid reactive periostitis ossificans4,6; however, observation and early range-of-motion exercises were successful in the case of our patient. Florid reactive periostitis ossificans shows little tendency toward progression and local recurrence13. This lesion probably is a reactive process rather than a true neoplasm2. Awareness of this lesion and careful radiographic and histological evaluation are needed in order to avoid misdiagnosis and unnecessary ablative operations.
*Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received, but are directed solely to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was the Henry L. Guenther Foundation.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Case Report
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |