Vertebral osteoblastoma: are radiologic structural changes necessary for diagnosis?

Imaging

Vertebral Osteoblastoma: Are Radiologic Structural Changes Necessary for Diagnosis? ¨ mu Toygun Orbay, M.D.,* O ¨ r Ataog ˇlu, M.D.,† E. Turgut Talı, M.D.,‡ Memduh Kaymaz, M.D.,* and Hızır Alp, M.D.* Departments of *Neurosurgery, †Pathology, and ‡Radiology, Gazi University Medical School, Ankara, Turkey

Orbay T, Ataog ˇlu O, Tali ET, Kaymaz M, Alp H. Vertebral osteoblastoma: are radiologic structural changes necessary for diagnosis? Surg Neurol 1999;51:426 –9. BACKGROUND

A case of osteoblastoma localized at the pedicle of the 10th thoracic vertebra is presented. CASE DESCRIPTION

The patient complained of nocturnal back pain not relieved by salycilates, a typical symptom of osteoblastoma. Bone scintigraphy showed a lower thoracic focus of increased osteoblastic activity; however, X-rays, computed tomography, and magnetic resonance images (MRI) were within normal limits, showing only obscure changes that were also noted in the rest of the spine. Repeat MRI with contrast revealed a focal enhancement. After pediculectomy, histopathologic examination confirmed the diagnosis of osteoblastoma. Fifteen months postoperatively, the patient is symptom-free. CONCLUSION

Our case demonstrates that some cases of osteoblastoma may not have the classical radiological appearance. Although non-contrast computed tomography and T1weighted MRI are mildly positive in some instances, osteoblastoma is best visualized on MRI with gadolinium. Like any other neoplasm, osteoblastoma should be detected and removed early, before it can cause structural bony changes. © 1999 by Elsevier Science Inc. KEY WORDS

Computed tomography, magnetic resonance imaging, nuclear bone scintigraphy, osteoblastoma, surgery.

steoblastoma is a relatively rare bone neoplasm, constituting approximately 1% of primary bone lesions [10]. It occurs primarily in the first two decades of life; the male/female ratio is 2:1 [4,10,16]. Although it is quite similar to osteoid osteoma both clinically and histopathologically, os-

O

Address reprint requests to: Dr. Toygun Orbay, 14. Sokak No. 7, Mutluko ¨ y, Ankara 06530 Turkey. Received April 22, 1996; accepted August 15, 1997. 0090-3019/99/$–see front matter PII S0090-3019(98)00103-7

teoblastoma is characterized by its larger size and lack of reactive perifocal bone formation. Its distinguishing symptom is pain that is not relieved by aspirin. Typically, the diagnosis is established by a focal increase in osteoblastic activity on bone scintography or by direct visualization on X-ray or computed tomography (CT) studies. The case presented is unique as only the postcontrast magnetic resonance (MR) examination confirmed the diagnosis.

Case Report A 19-year-old female patient was admitted with a 3-year history of intense back pain localized in the lower thoracic region. The pain was usually nocturnal, and the patient had not benefited from medication. She had been investigated at various medical centers, but no pathology was revealed by laboratory or radiological tests. She was sent to a psychiatry clinic because she had had suicidal thoughts; she was then referred to our clinic because her clinical picture suggested an organic cause. Neurologic examination was normal, as were laboratory studies and plain X-rays; however, a bone scan demonstrated focal increased osteoblastic activity involving the left T10 or T11 pedicle (Figure 1). CT scan demonstrated a slight focal hyperdensity (Figure 2) involving the left T10 pedicle. T1-weighted MRI showed mild hypointensity (Figure 3); T2-weighted images were normal. Highresolution MR with gadolinium-DTPA using fatsupressed images clearly showed asymmetrical contrast enhancement of the left T10 pedicle and transverse process (Figure 4). The tumor was removed via a left T10-11 costo© 1999 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

Vertebral Osteoblastoma

Surg Neurol 427 1999;51:426 –9

3

Axial T1-weighted shows a mild hypointensity of the left T10 pedicle.

scan obtained 15 months postoperatively showed no residual tumor.

Histopathology Bone scintigraphy showing focal increased osteoblastic activity in the T10-11 region (arrow).

transversectomy combined with a T10 transpedicular approach. At surgery, the bone was softened, but otherwise appeared normal. Postoperatively, the patient was asymptomatic; there was no increase of neurological dysfunction and no residual pain or discomfort. Because the tumor was totally removed, no further therapy was considered. CT

The specimen showed large mineralized bone structures reminiscent of the mosaic pattern in Paget’s disease. There were also areas rich in osteoid and newly formed bone, alternating with areas containing the mineralized bone structures described above. Dilated capillary vessels were seen in the loose connective tissue stroma. No central nidus was seen. There was no significant atypia or mitotic activity in the osteoblasts or in the fibroblastic cells of the stroma. Several multinuclear giant cells were noted in areas adjacent to the bony

CT of the 10th thoracic vertebra with a hyperdense 2 area involving the left pedicle. Note the absence of bony destruction.

Axial postcontrast fat-suppressed T1-weighted image demonstrates contrast enhancement of the left pedicle and transverse process (arrowheads) of T10, with no obvious structural change.

1

4

428 Surg Neurol 1999;51:426 –9

Histopathological appearance of the lesion. Areas containing osteoid (O) and large mineralized bone (M) structures in a loose fibrous stroma are observed. Numerous multinuclear cells are also present (arrowheads). Hematoxylin & eosin 3100.

5

structures. There were no appreciable epithelioid osteoblasts (Figure 5). In this case, the main differential diagnosis was between osteoid osteoma and osteoblastoma. Osteoid osteoma has little stromal reaction and few multinucleated giant cells [7]. A distinguishing feature of osteoblastoma is the absence of a central nidus. However, there are cases in which it is not possible to make a clear-cut distinction between the two on the basis of histology [6]. Schajowicz and Lemos considered both tumors to be osteoblastomas [15]. In our case, the histopathological findings and clinical history were more consistent with osteoblastoma. Osteoblastoma must also be differentiated from osteosarcoma. We did not observe the mitotic figures, significant cellular atypia, thick osteoid and woven bone, and vascularized stroma that are typical of osteosarcoma [7]. On the other hand, aggressive osteoblastoma has typical osteoblastoma areas and epitheloid-like structures formed by osteoblasts with large cytoplasm; these were not seen in our case [5].

Discussion Although osteoblastoma is a relatively rare entity, its clinical, radiological, and histopathological features are well defined. The most prominent symptom is a dull, localized, and gradually increasing pain [8]. Nocturnal pain is considered a major characteristic of osteoblastoma. However, pain is not restricted to the nighttime hours; it occurs throughout the day [8]. Unlike that of osteoid osteoma, the

Orbay et al

pain associated with osteoblastoma usually does not respond to salicylates [8]. Neurologic deficits have been reported in 25–50% of patients [12]. Osteoblastoma most often originates in the posterior vertebral elements (i.e., the pedicles and laminae) and occasionally extends into the vertebral bodies themselves [8,10,12]. The radiological changes associated with vertebral osteoblastoma include a lesion larger than 1 cm with variable margins (well-defined, intermediate, or ill-defined), central calcification, reactive sclerosis and cortical expansion of adjacent normal bone, periosteal bone formation, and demarcation by a thin bone shell [8,12]. We do not believe that the size of the lesion should be considered a diagnostic criterion, since the diagnosis of osteoblastoma is often delayed [13]. CT and MRI findings have been reported in detail elsewhere [8,16]. These findings depend on the morphological changes in the involved bone. On reviewing the literature, we were unable to find any case in which the bony structure was unaffected. It is well known that nuclear bone scanning has a very high sensitivity in detecting bony neoplasms including osteoblastomas. Mohan et al stated that if an area of sclerosis is observed in a patient with back pain, bone scanning is the study of choice; if it is positive, CT and MRI studies must be obtained [11]. However, a problem can still arise if the patient has a clinical picture of osteoblastoma (or osteoid osteoma) with a positive bone scan, but has non-specific findings on X-ray, CT, and MRI. Our limited experience may indicate that: 1. A history of intense localized pain, whether or not it is nocturnal, should be thoroughly investigated, even if there are no radiological changes. 2. Surgery is warranted when neuroradiological studies reveal (a) mild CT and T1 MR changes with or without changes on plain X-ray; (b) marked abnormalities on bone scan; or (c) marked abnormalities on contrast-enhanced MR (especially on fat-suppressed images). 3. Radical removal of the lesion should be attempted, because it has been shown that incomplete removal of an osteoblastoma may result in recurrence, although incomplete curettage alone may result in a cure in a high percentage of patients [8]. It should also be kept in mind that malignant transformation and aggressive behavior of osteoblastoma has been demonstrated [1,10,17]. Radiotherapy can be used in cases of aggressive tumor, but to eliminate the risk of postradiation sarcoma, it should be avoided when radical surgery is possible [2,14,15].

Vertebral Osteoblastoma

Likewise, chemotherapy may be useful in selected patients, but only if an aggressive tumor cannot be surgically removed or if it recurs [3]. REFERENCES 1. Adler M, Hnatuk L, Mock D, Freeman JL. Aggressive osteoblastoma of the temporal bone: a case report. J Otolaryngol 1990;19:307–10. 2. Boriani S, Capanna R, Donati D, Levine A, Picci P, Savini R. Osteoblastoma of the spine. Clin Orthop 1992;278:37– 45. 3. Camitta B, Wells R, Segura A, Unni KK, Murray K. Osteoblastoma response to chemotherapy. Cancer 1990;68:999 –1009. 4. DiLorenzo N, Nardi P, Ciapetta P, Fortuna A. Benign tumors and tumorlike conditions of the spine: Radiological features, treatment, and results. Surg Neurol 1986;25:449 –56. 5. Dorfman HD, Weiss SW. Borderline osteoblastic tumors: Problems in differential diagnosis of aggressive osteoblastoma and low-grade osteosarcoma. Semin Diag Pathol 1984;1:215–34. 6. Greenspan A. Benign bone-forming lesions: Osteoma, osteoid osteoma, and osteoblastoma: clinical, imaging, pathologic, and differential considerations. Skeletal Radiol 1993;22:485–500. 7. Huvos AG. Bone tumors. Diagnosis, treatment, and prognosis. 2nd ed. Philadelphia: W. B. Saunders, 1991: 67– 83. 8. Kroon HM, Schurmans J. Osteoblastoma: clinical and radiologic findings in 98 new cases. Radiology 1990; 175:783–90. 9. Lichtenstein L. Bone tumors. 4th ed. St. Louis: Mosby, 1972:103–20. 10. Lucas DR, Unni KK, McLeod RA, O’Connor MI, Sim FH. Osteoblastoma: clinicopathologic study of 306 cases. Hum Pathol 1994;5:117–34. 11. Mohan V, Sabri T, Marklund T, Sayed M, Gupta RP. Clinicoradiological diagnosis of benign osteoblastoma of the spine in children. Arch Orthop Trauma Surg 1991;110:260 – 4. 12. Nyugen VD, Hersh M. A rare bone tumor in an unusual location: osteoblastoma of the vertebral body. Comput Med Imaging Graph 1992;16:11–16. 13. Paige ML, Michael AS, Brodin A. Case report 647: benign osteoblastoma causing spinal cord compression and spastic paresis. Skeletal Radiol 1991;20:54 –7. 14. Seki T, Fukuda H, Ishii Y, Hanaoka H, Yatabe S, Tanako M. Malignant transformation of benign osteoblastoma: a case report. J Bone Joint Surg 1975;57A: 424 – 6. 15. Schajowicz F, Lemos C. Osteoid osteoma and osteoblastoma: closely related entities of osteoblastic derivation. Acta Orthop Scand 1970;41:272–91. 16. Syklawer R, Osborn RE, Kerber CW, Glass RF. Magnetic resonance imaging of vertebral osteoblastoma: a report of two cases. Surg Neurol 1990;34:421– 6. 17. Vara-Thorbeck R, Morales OI, Rosell J, Gomez M. Benign osteoblastoma of vertebral column and skull: report of two cases. Zentralb Neurochir 1990;51: 216 – 8.

Surg Neurol 429 1999;51:426 –9

COMMENTARY

I agree clinically that sometimes it can be very difficult to differentiate osteoblastomas from osteoid osteomas. Osteoblastomas are usually greater than 1 cm in size; osteoid osteomas tend to be smaller. Also, osteoblastomas have a tendency to form a less sporadic but more expansile mass. Osteoblastomas tend to progress and may undergo malignant changes. The authors begin by stating that osteoblastomas are similar to osteoid osteomas, but the pain is not relieved by aspirin. Aspirin and salicylates can be effective for relieving the pain associated with osteoblastoma, but probably less so than with osteoid osteoma. However, the literature demonstrates that osteoblastoma has a history of back pain that increases in the evening and is relieved by aspirin or salicylates. In all regions of the spine, the posterior elements and pedicles are more frequently involved than the vertebral body. Pain and other symptoms usually persist for more than 12 months before the diagnosis is made. Some lesions of the thoracic spine will demonstrate neurologic involvement; whenever a patient presents with painful scoliosis (or a child with painful torticollis), we should keep this diagnosis in mind. Osteoblastomas, especially those in the cervical and thoracic spine, require complete therapeutic resolution; generally resection is efficacious and without recurrence. The aim of surgical treatment in every case should be tumor removal, decompression of the spinal cord, and restoration of vertebral stability. This is a very exciting presentation, and obviously the literature will provide different opinions. For some, the classic presentation is nocturnal pain alleviated with salicylates. Others state that salicylates are much less effective for osteoblastoma than for osteomas. The surgical goal should be complete excision of the tumor; some authors even advise the addition of radiation therapy. From a surgical standpoint, in any mobile part of the spine, complete excision of the tumor and stabilization with graft and instrumentation should be an obvious concern. Although osteoblastomas account for less than 1% of bone tumors, it should always be in mind as a diagnosis in cases of a painful, tender area of the spine, as well as the possibility of painful scoliosis or torticollis. Renato Iman ˜ a, M.D. Neurosurgeon Oak Brook, Illinois