Published online before print January 28, 2004, 10.1148/radiol.2303030226
(Radiology 2004;230:697-702.)
© RSNA, 2004
Primary Vertebral Osteosarcoma: Imaging Findings1
Hakan Ilaslan, MD2,
Murali Sundaram, MBBS, FRCR,
K. Krishnan Unni, MBBS and
Thomas C. Shives, MD
1 From the Departments of Radiology (H.I., M.S.), Pathology and Laboratory Medicine (K.K.U.), and Orthopedic Surgery (T.C.S.), Mayo Clinic, Ch2–290, 200 First St SW, Rochester, MN 55905. Received February 10, 2003; revision requested April 25; final revision received July 18; accepted August 22. Address correspondence to M.S. (e-mail: sundaram.murali@mayo.edu).
 |
ABSTRACT
|
|---|
PURPOSE: To evaluate patient age and sex and location and imaging appearances of primary vertebral osteosarcoma (PVOS) compared with histologic subtypes.
MATERIALS AND METHODS: Retrospective review (1915–2001) of imaging findings in patients with histologically proved primary osteosarcoma of vertebral column was performed. Two radiologists in consensus reviewed findings for location, origin site, matrix pattern, and spinal canal invasion and compared them with histologic subtypes. Radiation-induced, Paget, metastatic, and multifocal osteosarcoma were excluded.
RESULTS: Of 4,887 osteosarcoma cases, 198 (4%) were PVOS arising from vertebral column. There were 103 female and 95 male patients (age range, 8–80 years; median age, 34.5 years). Involvement included cervical (27 patients), thoracic (66 patients), lumbar (64 patients), and sacral (41 patients) spine. Adequate imaging findings were available in 69 patients, and involvement of two levels was seen in 12 (17%). In nonsacral spine, most tumors (44 cases) arose from posterior elements, with partial involvement of vertebral body. Lesions confined to vertebral body were less frequent (12 cases). Sacral tumors involved body and sacral ala. The most common histologic subtypes were osteoblastic (47 patients), chondroblastic (12 patients), telangiectatic (four patients), fibroblastic (four patients), small cell (one patient), and epithelioid (one patient). The majority (55 cases) demonstrated osteoid matrix mineralization; 17 showed marked mineralization. Five cases with marked mineralization were confined to vertebral body, with "ivory vertebra" appearance. Purely lytic pattern was seen in 14 (20%) cases. Lytic pattern was seen in four (100%) telangiectatic, three (75%) fibroblastic, three (25%) chondroblastic, three (6%) conventional osteoblastic, and one (100%) small-cell subtypes. Invasion of spinal canal was common (84% of cases). Appearance simulating osteoblastoma without soft-tissue mass was present (seven cases). Pathologic compression fractures were identified (seven patients).
CONCLUSION: This study provides age and sex distribution and location and imaging features in a large series of PVOS.
© RSNA, 2004
Index terms: Myelography, 30.122 • Osteosarcoma, 30.322 • Spine, CT, 30.1211 • Spine, MR, 30.1214 • Spine, radiography, 30.11, 30.122
|
INTRODUCTION
|
|---|
Osteosarcoma is the most common nonhematologic primary malignancy of bone and has been extensively documented in the appendicular skeleton (1,2). Vertebral osteosarcoma has been the subject of case reports and series ranging from 10 to 30 cases (3–5). Of these, we are aware of only one series (5) that was limited to primary vertebral osteosarcoma (PVOS), whereas others included secondary vertebral osteosarcoma that resulted from Paget disease and radiation (3,4). The largest series with review of imaging studies included 12 patients and was limited to radiographs (4). We are not aware of any series in which imaging findings were compared with the histologic types of osteosarcoma or in which conclusions were drawn on the basis of demographics and distribution of tumor in the axial skeleton. Thus, the purpose of our study was to evaluate patient age and sex and location and imaging appearances of PVOS compared with histologic subtypes.
|
MATERIALS AND METHODS
|
|---|
Subjects and Imaging Study Review
After obtaining approval from the institutional review board, one of the authors (H.I.) performed a retrospective review of pathology records of patients who received a diagnosis of osteosarcoma; this review included records of patients examined at our institution and records from consultation cases between 1915 and 2001. Consent was provided to review all records, with no patients having declined study of their records for research purposes. Patients with secondary vertebral osteosarcoma, that is, secondary to radiation or arising in Paget disease were excluded. Patients with known osteosarcoma in other anatomic areas were excluded.
Available images of PVOS were reviewed by two musculoskeletal radiologists (H.I. [5 years experience], M.S. [30 years experience]) in a nonblinded fashion, and findings were documented with consensus. Images evaluated were radiographs, computed tomographic (CT) scans, magnetic resonance (MR) images, and myelograms. These images were evaluated for location of lesion, site of origin, morphology (expansive or not), fracture, and multilevel involvement. The matrix of the lesion (mineralized or lytic) was evaluated from radiographs, CT scans, and myelograms. Mineralization was subjectively evaluated as marked or subtle. Spinal canal invasion was evaluated on myelograms, CT scans, and MR images. CT scans and MR images were further evaluated for surface lesions, fluid-fluid levels, and expansive lesions without soft-tissue mass. The images were deemed adequate if they were able to depict one or more of the features being examined.
Histologic Findings
The histologic subtypes documented at another review (K.K.U. [35 years experience]) of all 198 cases in this study were osteoblastic, chondroblastic, telangiectatic, fibroblastic, small cell, epithelioid, low-grade intraosseous, and parosteal (1). Prior to this other review, histologic slides had been periodically reviewed by the same author as just mentioned and a retired radiologist with 45 years experience for consistency with current histologic concepts and accuracy. In the majority of cases, open surgical biopsy was performed; in a smaller number, percutaneous biopsy was performed.
Comparisons
Age and sex of patients were determined from consultation and medical records by one author (H.I.), who compared age and sex with location of lesion in all cases. Age and sex were determined separately for the cases from our institution and for those from consultation and were compared with one another.
Statistical Analysis
We report the basic descriptive statistics for age and sex. Wilcoxon rank sum tests were used to test the differences in the median ages for female versus male patients and also for patients seen at our institution versus patients in the consultation group. The differences in the proportions of male patients and female patients who were seen at our institution versus the differences in the proportions of those in the consultation group, as well as the median ages of patients according to sex in the respective groups, were evaluated by using the 
2 test.
|
RESULTS
|
|---|
Subjects and Imaging Studies
There were a total of 4,887 osteosarcoma cases: 1,861 cases from our institution and 3,026 cases from consultation. One hundred ninety-eight (4%) primary osteosarcomas originated in the vertebral column. For the PVOS group, 143 patients were in the consultation group, and 55 patients were seen at our institution. Thirty-five patients with secondary vertebral osteosarcoma who were seen at our institution were excluded. Thirty-two of these had radiation-induced osteosarcoma; three had Paget sarcoma.
In the PVOS group, there were 103 female patients and 95 male patients. The age of the patients ranged from 8 to 80 years (median age, 34.5 years). In the consultation group, there were 67 male patients and 76 female patients; among these, age information was available in 64 male patients (age range, 8–75 years; median age, 27 years) and in 71 female patients (age range, 9–79 years; median age, 41 years). Twenty-eight male patients whose ages ranged from 14 to 80 years (median age, 29.5 years) and 27 female patients whose ages ranged from 10 to 80 years (median age, 42 years) were seen at our institution. The overall median age for the consultation group was 35 years, and the median age for the patients seen at our institution was 34 years. The aforementioned data about age in both groups were analyzed by one author (M.S.). Imaging study findings in 69 patients were deemed suitable for evaluation of the aforementioned imaging characteristics indicated in Materials and Methods and were thus included in our study, except that all 198 were evaluated for vertebral location. The studies consisted of 45 radiographic, 26 CT, and 16 MR imaging examinations. The CT and MR imaging studies were performed in 38 patients; also, seven patients underwent myelography. CT and MR imaging studies were not uniform in regard to imaging parameters and were performed with a wide range of imaging units belonging to different generations.
Imaging Findings and Comparisons
Location of vertebral involvement was as follows: cervical spine, 27 (13.6%) patients; thoracic spine, 66 (33.3%) patients; lumbar spine, 64 (32.3%) patients; and sacral spine, 41 (20.7%) patients. In the 69 patients with suitable images, most tumors (44 [79%] of 56) in the nonsacral spine arose from the posterior elements, with partial involvement of the vertebral body. Lesions confined to the vertebral body were seen less frequently (12 [21%] of 56 cases). All 13 (100%) sacral tumors with available imaging showed involvement of the body and sacral ala in all (Fig 1). The most common histologic type in the 69 patients with adequate imaging was osteoblastic (47 patients); others included chondroblastic (12 patients), telangiectatic (four patients), fibroblastic (four patients), small-cell (one patient), and epithelioid (one patient) subtypes. There were no low-grade intraosseous, parosteal, or other surface osteosarcomas.
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Transverse CT scan in 32-year-old man shows destructive lesion of sacral body and left sacral ala caused by osteoblastic osteosarcoma, with large partially mineralized soft-tissue mass (arrows).
|
|
In the majority (n = 55) of cases, radiographs and CT scans demonstrated mineralized matrix (Figs 1, 2); in 17 of them, marked mineralization was observed (Fig 2b). Five (7%) of 69 cases with marked mineralization were confined to the vertebral body, with a typical ivory vertebra appearance (Fig 2c). A purely lytic pattern was less common and was seen in 14 (20%) cases. A lytic pattern according to histologic subtypes was as follows: telangiectatic, four (100%) of four (three radiographs, one MR image) (Fig 3); fibroblastic, three (75%) of four (three radiographs, two CT scans, one MR image); chondroblastic, three (25%) of 12 (three radiographs, one CT scan); conventional osteoblastic, three (6%) of 47 (three radiographs, one CT scan); and small-cell, one (100%) of one (one set of radiographs, one CT scan, and one MR image).
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. (a) Transverse CT scan of thoracic vertebra in 46-year-old man shows more diffusely mineralized mass (arrow) arising from and associated with some destruction of the right transverse process and pedicle caused by chondroblastic osteosarcoma associated with intraspinal extension. (b) Lateral radiograph of cervical spine in 34-year-old woman shows intense sclerosis (*) of posterior elements, body, and dens of C2 caused by osteoblastic osteosarcoma that completely obscures normal bony outlines and extends superiorly to the arch of C1, which appears intact. (c) Lateral radiograph of lumbar spine in 37-year-old man shows osteoblastic osteosarcoma with "ivory vertebra" (*) appearance at L4 involving body and pedicles.
|
|
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. (a) Transverse CT scan of thoracic vertebra in 46-year-old man shows more diffusely mineralized mass (arrow) arising from and associated with some destruction of the right transverse process and pedicle caused by chondroblastic osteosarcoma associated with intraspinal extension. (b) Lateral radiograph of cervical spine in 34-year-old woman shows intense sclerosis (*) of posterior elements, body, and dens of C2 caused by osteoblastic osteosarcoma that completely obscures normal bony outlines and extends superiorly to the arch of C1, which appears intact. (c) Lateral radiograph of lumbar spine in 37-year-old man shows osteoblastic osteosarcoma with "ivory vertebra" (*) appearance at L4 involving body and pedicles.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. (a) Transverse CT scan of thoracic vertebra in 46-year-old man shows more diffusely mineralized mass (arrow) arising from and associated with some destruction of the right transverse process and pedicle caused by chondroblastic osteosarcoma associated with intraspinal extension. (b) Lateral radiograph of cervical spine in 34-year-old woman shows intense sclerosis (*) of posterior elements, body, and dens of C2 caused by osteoblastic osteosarcoma that completely obscures normal bony outlines and extends superiorly to the arch of C1, which appears intact. (c) Lateral radiograph of lumbar spine in 37-year-old man shows osteoblastic osteosarcoma with "ivory vertebra" (*) appearance at L4 involving body and pedicles.
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Telangiectatic osteosarcoma in 37-year-old woman. (a) Lateral radiograph of spine shows destruction of posterior half of T12 (*), with cortical destruction and complete disappearance of posterior elements without sclerosis of bone or identifiable mineral in lesion. Black artifact in posterior portion of vertebral body could not be erased. (b) Sagittal T2-weighted MR image of spine (repetition time msec/echo time msec, 4,290/92) shows posterior elements replaced by expansive mass containing large and small fluid-fluid levels (arrows). Posterior half of vertebral body (arrowhead) shows areas of abnormal signal intensity caused by tumor.
|
|
View larger version (187K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Telangiectatic osteosarcoma in 37-year-old woman. (a) Lateral radiograph of spine shows destruction of posterior half of T12 (*), with cortical destruction and complete disappearance of posterior elements without sclerosis of bone or identifiable mineral in lesion. Black artifact in posterior portion of vertebral body could not be erased. (b) Sagittal T2-weighted MR image of spine (repetition time msec/echo time msec, 4,290/92) shows posterior elements replaced by expansive mass containing large and small fluid-fluid levels (arrows). Posterior half of vertebral body (arrowhead) shows areas of abnormal signal intensity caused by tumor.
|
|
Invasion of the spinal canal secondary to a soft-tissue mass was common and was seen on images (18 CT scans, 10 MR images, four combined CT scans and MR images, six myelograms) in 84% (38 of 45) of patients. The finding of a lack of a soft-tissue mass was based on four CT scans, two MR images, and one myelogram. Fluid-fluid levels were seen in one patient with a telangiectatic osteosarcoma (Fig 3). Involvement of two vertebral levels was seen in 12 (17%) cases (Fig 4). The locations of the involvement in these 12 cases were as follows: L5 through S1, five cases; T12 through L1, two cases; and C3–4, C6–7, T7–8, L1–2, and L4–5, one case each. An appearance of an expansive lesion that simulates osteoblastoma without a soft-tissue mass was present in seven cases (Fig 5). Pathologic compression fractures were identified in seven patients (Fig 6).
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Lateral radiograph of cervical spine in 46-year-old woman with contiguous multilevel osteoblastic osteosarcoma, with sclerosis of the anterior margins of C3-4 and associated partial collapse and destruction.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. (a) Transverse CT scan of vertebra in 20-year-old man shows expansive mass (arrows) with an outer shell of bone and no soft-tissue mass. Entire left half of posterior elements and posterior body were involved. More than 50% of lesion was mineralized, and dense bone invaded spinal canal. This mass was a grade 3 osteoblastic osteosarcoma. (b) Transverse CT scan of vertebra in 30-year-old woman shows expansive lesion (arrows), less expansive than that in a, with no soft-tissue mass and clearly defined margins. Entire lesion was ossified; portions of it occupied spinal canal; and it was in partial contact with the psoas muscle superolaterally. Fat plane was evident inferolaterally between psoas muscle and ossified tumor. This mass was a grade 3 osteoblastic osteosarcoma.
|
|
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. (a) Transverse CT scan of vertebra in 20-year-old man shows expansive mass (arrows) with an outer shell of bone and no soft-tissue mass. Entire left half of posterior elements and posterior body were involved. More than 50% of lesion was mineralized, and dense bone invaded spinal canal. This mass was a grade 3 osteoblastic osteosarcoma. (b) Transverse CT scan of vertebra in 30-year-old woman shows expansive lesion (arrows), less expansive than that in a, with no soft-tissue mass and clearly defined margins. Entire lesion was ossified; portions of it occupied spinal canal; and it was in partial contact with the psoas muscle superolaterally. Fat plane was evident inferolaterally between psoas muscle and ossified tumor. This mass was a grade 3 osteoblastic osteosarcoma.
|
|
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. (a) Lateral radiograph of cervical spine in 23-year-old woman with vertebra plana (*) at C6. Body is sclerotic, with bone production anteriorly secondary to osteoblastic osteosarcoma. (b) Delayed contrast-enhanced sagittal T1-weighted MR image (600/14) of cervical spine shows area of diffuse low signal intensity in collapsed C6 (*), with extradural mass displacing the cord posteriorly. Area of slightly increased signal intensity at C5 and C7 is of unknown clinical significance. There had been no previous radiation therapy. (c) Sagittal T2-weighted MR image (3,816/90) of cervical spine shows area of low signal intensity (*), as expected with a sclerotic lesion, and large posterior mass compressing spinal cord.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. (a) Lateral radiograph of cervical spine in 23-year-old woman with vertebra plana (*) at C6. Body is sclerotic, with bone production anteriorly secondary to osteoblastic osteosarcoma. (b) Delayed contrast-enhanced sagittal T1-weighted MR image (600/14) of cervical spine shows area of diffuse low signal intensity in collapsed C6 (*), with extradural mass displacing the cord posteriorly. Area of slightly increased signal intensity at C5 and C7 is of unknown clinical significance. There had been no previous radiation therapy. (c) Sagittal T2-weighted MR image (3,816/90) of cervical spine shows area of low signal intensity (*), as expected with a sclerotic lesion, and large posterior mass compressing spinal cord.
|
|
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. (a) Lateral radiograph of cervical spine in 23-year-old woman with vertebra plana (*) at C6. Body is sclerotic, with bone production anteriorly secondary to osteoblastic osteosarcoma. (b) Delayed contrast-enhanced sagittal T1-weighted MR image (600/14) of cervical spine shows area of diffuse low signal intensity in collapsed C6 (*), with extradural mass displacing the cord posteriorly. Area of slightly increased signal intensity at C5 and C7 is of unknown clinical significance. There had been no previous radiation therapy. (c) Sagittal T2-weighted MR image (3,816/90) of cervical spine shows area of low signal intensity (*), as expected with a sclerotic lesion, and large posterior mass compressing spinal cord.
|
|
The median, minimum, and maximum ages for male patients were 27.5, 8, and 80 years and for female patients were 41.5, 9, and 80 years. Though the difference in age was not statistically significant (P = .06), it may be important clinically. The median, minimum, and maximum ages of patients seen at our institution were 34, 10, and 80 years and those of patients in the consultation group were 35, 8, and 79 years. A P value of .50 indicated that there was no statistically significant difference between the median age of patients seen at our institution and that of patients in the consultation group. The 2 test showed no significant differences in the proportions of male patients (51% vs 47%) and female patients (49% vs 53%) between the two groups (P = .61, Mayo Clinic vs consultation group). The median ages of male patients in the consultation group and the group at our institution were 27 and 29.5 years, respectively (P = .32). The median ages for female patients in the consultation and institution groups were 41 and 42 years, respectively (P = .98). There was no significant difference in median age with respect to sex between the two groups.
|
DISCUSSION
|
|---|
The frequency of primary osteosarcoma in the axial skeleton was 4% on the basis of our study of 4,887 cases of all osteosarcomas. The mean age of incidence in the 4th decade is of interest in that it is 2 decades later than the mean age of its appendicular counterpart, which is dissimilar to vertebral chondrosarcoma (6,7), where no striking age difference was noted between axial and appendicular tumors. The age distribution of osteosarcoma is bimodal; a second smaller peak in incidence of the tumor with respect to age occurs in patients older than 50 years, and this peak is associated with a relative increase in pelvic and craniofacial involvement and a decrease in involvement of extremity bones (8). The median age of vertebral involvement would appear to be between the ages for appendicular and pelvic involvement with osteosarcoma. The median ages of the patients in our consultation and institution groups were not significantly different. The slightly increased number of female patients is at variance from the number for osteosarcoma overall, with 103 female patients and 95 male patients in our study group. The thoracic and lumbar segments (130 cases) were equally involved, followed by the sacrum and the cervical vertebral column. The frequency of thoracic involvement with osteosarcoma runs counter to previous observations that the thoracic vertebra is a rare site and lumbosacral predilection is favored (9,10). A review of 41 cases from 1962 to 1984 revealed the thoracic vertebra to be the least involved site (9).
On the basis that only slightly more than one-third of our cases had imaging findings suitable for characterization of lesion origin and matrix, it was found that in 44 (79%) of 56 cases in the mobile spine, the tumor arose in the posterior elements with partial body involvement. This pattern is in keeping with the most frequently encountered aggressive benign vertebral lesions (aneurysmal bone cyst, osteoblastoma) and chondrosarcoma (6,11), but again this information contradicts previous implications about vertebral site of origin (10,12). The 17% involvement of two levels would appear to be higher than it is in other vertebral tumors, benign and malignant.
As would be expected with a high-grade tumor, spinal canal invasion was seen in 84% of cases. In 10% of cases without a soft-tissue mass, the tumor was indistinguishable from osteoblastoma, and some were markedly mineralized. It is not surprising that a tumor with the hallmark of osteoid production and mineralization should produce an appearance described as an ivory vertebra, a feature most often associated with sclerotic metastasis, lymphoma, and Paget disease. This finding was present in 7% of cases, and we have identified one other such description in the English-language literature (5). Intense sclerosis of a hemivertebra was the most common radiographic appearance in an earlier series (3).
Of note in the histologic subtypes was the absence of parosteal osteosarcoma (in keeping with the absence of a surface lesion at imaging) and low-grade intraosseous osteosarcoma, although the rare small-cell and epithelioid osteosarcomas were encountered. There is no uniformity among the current classification systems of osteosarcoma, and categorizing them into lytic and sclerotic subtypes is not considered to be of practical importance (1,2), with the exception of telangiectatic osteosarcoma (1). A purely osteolytic lesion was noted in 20% of our cases, which included 75% of the fibroblastic subtype cases and all of the telangiectatic osteosarcoma cases. Six percent of the conventional osteoblastic type were also osteolytic at imaging. Telangiectatic osteosarcoma is a diagnosis that can be rendered if the radiograph shows a predominantly lytic lesion. If any substantial sclerosis is present, this diagnosis usually can be ruled out (1). As already noted, all four telangiectatic osteosarcoma cases were osteolytic. In the only case in which cross-sectional imaging was performed, fluid-fluid levels were demonstrated on an MR image. The fluid-fluid level is a well-recognized feature in the appendicular skeleton, although it is more frequently associated with an aneurysmal bone cyst.
There were limitations to our study. Cross-sectional imaging, which today is considered essential for staging and treatment planning for vertebral lesions, has been in existence for only a little less than 3 decades. Our study group spans 9 decades, and the findings from cross-sectional imaging were limited to only 38 patients. The quality of advanced imaging that spanned 3 decades was inevitably nonuniform, and its influence on clinical treatment was not assessed. The study permitted assessment of spinal canal invasion that was based on imaging, although this was not compared with surgical or autopsy findings. In this study, we made no attempt to compare prognosis and survival of patients with PVOS with the progress and survival of patients who have its appendicular counterpart. In this study, we provide demographics, lesion distribution, and range of imaging appearances of PVOS that have been compared with histologic subtypes from what we believe is the largest series about the subject that is based on a search of the world literature.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank Linda Greene for her assistance in manuscript preparation and Jayawant Mandrekar, PhD, for providing statistical analysis.
|
FOOTNOTES
|
|---|
2 Current address: Department of Radiology, Cleveland Clinic, Ohio. 
Abbreviation: PVOS = primary vertebral osteosarcoma
Author contributions: Guarantor of integrity of entire study, H.I.; study concepts and design, all authors; literature research, H.I.; clinical studies, K.K.U., T.C.S.; data acquisition and analysis/interpretation, all authors; statistical analysis, M.S., H.I.; manuscript preparation, definition of intellectual content, revision/review, and final version approval, all authors; manuscript editing, M.S.
|
REFERENCES
|
|---|
- Unni KK. Osteosarcoma. Dahlin’s bone tumors: general aspects and data on 11,087 cases 5th ed. Philadelphia, Pa: Lippincott-Raven, 1996; 143-183.
- Dorfman HD, Czerniak B. Osteosarcoma In: Bone tumors. St Louis, Mo: Mosby, 1998; 128-252.
- Barwick KW, Huvos AG, Smith J. Primary osteogenic sarcoma of the vertebral column: a clinicopathologic correlation of ten patients. Cancer 1980; 46:595-604.[CrossRef][Medline]
- Shives TC, Dahlin DC, Sim FH, Pritchard DJ, Earle JD. Osteosarcoma of the spine. J Bone Joint Surg Am 1986; 68:660-668.[Abstract/Free Full Text]
- Tigani D, Pignatti G, Picci P, Savini R, Campanacci M. Vertebral osteosarcoma. Ital J Orthop Traumatol 1988; 14:5-13.[Medline]
- Shives TC, McCleod RA, Unni KK, Schray MF. Chondrosarcoma of the spine. J Bone Joint Surg Am 1989; 71:1158-1165.[Abstract/Free Full Text]
- York JE, Berk RH, Fuller GN, et al. Chondrosarcoma of the spine: 1954 to 1997. J Neurosurg 1999; 90:73-78.[Medline]
- Dorfman HD, Czerniak B, Kotz R, Vanel D, Park YK, Unni KK. WHO classification of tumours of bone: introduction. In: Fletcher CD, Unni KK, Mertens F, eds. WHO classification of tumours: pathology and genetics of tumors of soft tissue and bone. Lyon, France: International Agency for Research on Cancer, 2002; 227-232.
- Miller TT, Abdelwahab IF, Hermann G, Morgello S. Vertebral osteosarcoma: case report 735. Skeletal Radiol 1992; 21:277-279.[Medline]
- Murphey MD, Andrews CL, Flemming DJ, Temple HT, Smith WS, Smirniotopoulos JG. Primary tumors of the spine: radiologic pathologic correlation. RadioGraphics 1996; 16:1131-1158.[Abstract]
- Rosenberg AE, Nielsen GP, Fletcher JA. Aneurysmal bone cyst. In: Fletcher CD, Unni KK, Mertens F, eds. WHO classification of tumours: pathology and genetics of tumors of soft tissue and bone. Lyon, France: International Agency for Research on Cancer, 2002; 338-339.
- A 14-year-old with abnormal bones and a sacral mass. Massachusetts General Hospital Case Records, case 29–2001. N Engl J Med 2001; 345:903-908.[Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
M. H. Rodallec, A. Feydy, F. Larousserie, P. Anract, R. Campagna, A. Babinet, M. Zins, and J.-L. Drape
Diagnostic Imaging of Solitary Tumors of the Spine: What to Do and Say
RadioGraphics,
July 1, 2008;
28(4):
1019 - 1041.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. P. Dormans and L. Moroz
Infection and Tumors of the Spine in Children
J. Bone Joint Surg. Am.,
February 1, 2007;
89(suppl_1):
79 - 97.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
