Surgical Neurology 71 (2009) 234 – 237

www.surgicalneurology-online.com

Neoplasm
Spinal cord glioblastoma multiforme of conus medullaris masquerading as
high lumbar disk herniation
Won-Chul Choi, MD a , June Ho Lee, MD b , Sang-Ho Lee, MD, PhD b,⁎
a
Department of Neurosurgery, Daegu Wooridul Spine Hospital, Daegu 157-822, Korea
b
Department of Neurosurgery, Wooridul Spine Hospital, Seoul 135-100, Korea
Received 28 March 2007; accepted 23 July 2007

Abstract Background: A case of primary intramedullary glioblastoma occurring at conus medullaris is

presented, which was initially undetected and misdiagnosed as L1 through L2 disk herniation.
Case Description: After the performance of L1 through L2 discectomy, the patient's initial painful
symptom did not subside. The patient then developed a progressive paraparesis. The outside
surgeons only predicted the possibility of reherniation or hematoma collection at the former operated
level as the diagnosis for the paraparesis. After the patient was transferred to our hospital and after a
thorough review of initial and follow-up magnetic resonance imaging with enhancement before
second operation, an enlarged conus medullaris with mixed signal intensity was noted without
evidence of re-herniation or hematoma collection. The patient underwent reoperation with biopsy of
the intramedullary lesion, which was histologically confirmed as primary glioblastoma.
Conclusion: Careful neurologic and radiologic evaluations should precede before any decision of
surgical intervention as intramedullary and degenerative pathologies may coexist in the upper
lumbar spine.
© 2009 Elsevier Inc. All rights reserved.
Keywords: Spinal cord; Glioblastoma; Whole spine scan MRI

1. Introduction In this report, we present such a case that was

misdiagnosed as L1 through L2 disk herniation initially.
Glioblastoma as a primary occurrence at the spinal cord is
We learned a valuable lesson from this case. When a
a rare disease entity [3,4], representing only about 7.5% of all
patient's initial diagnosis and the neurologic dysfunction are
intramedullary glioma [5,7]. Moreover, it has the predilec-
mismatched, we proceed with further detailed study such as
tion of development at the cervical or cervicothoracic region
enhanced MRI of whole spinal axis.
in more than 60% [2,5,7,10], with rare incidence at conus
medullaris [6]. Therefore, it would be extremely unusual that
a patient initially presents with severe back pain only with 2. Case report
degenerative disk disease at the upper lumbar level and a
2.1. Preoperative neurologic status and radiologic findings
glioblastoma of the conus medullaris.
A 46-year-old male patient was transferred to our
institution with severe back pain and progressive weak-
ness of both lower extremities. He was not able to
maintain a supine position because of the severe pain.
Abbreviations: CT, Computed tomography; MRI, Magnetic resonance
imaging. Two months before, the patient underwent L1 through L2
⁎ Corresponding author. Tel.: +82 2 513 8151; fax: +82 2 513 8146. laminectomy and discectomy at a local clinic after being
E-mail address: shlee@wooridul.co.kr (S.-H. Lee). diagnosed as having a severe disk herniation at the L1
0090-3019/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.surneu.2007.07.059
 W.-C. Choi et al. / Surgical Neurology 71 (2009) 234–237 235

Fig. 1. Sagittal T1-weighted (A) and T2-weighted (B) MR image before first operation showing L1/L2 disk extrusion and conus enlargement.

through L2 level. Before the initial operation, he only had enlarged conus medullaris (Fig. 1). A plain follow-up
back pain and left lower extremity radiculopathy without thoracic MRI taken 2 months after the first operation at a
motor and sensory deficit. However, after the initial local clinic demonstrated a marked cord enlargement at
operation, his symptoms have not improved. He then thoracolumbar junction when compared with the initial MRI
developed a progressive paraparesis. finding (Fig. 2). Suspecting a tumorous condition at the
On presentation to our hospital, his neurologic examina- conus, an enhanced MRI and myelography were performed
tion results demonstrated paraparesis (right/left: grades IV at our institution. This revealed no definite enhancement of
and II, respectively) and hypesthesia below L1 sensory the conus but did demonstrate almost totally blocked
dermatome. The initial MRI taken before first operation cerebrospinal fluid flow at the T12 through L1 level. The
showed L1 through L2 disk herniation and somewhat whole-spine MRI with enhancement and brain CT did not

Fig. 2. Thoracic image after discectomy. A: T1-weighed sagittal image depicting enhanced mass lesion at conus medullaris (arrow). B: T2-weighted sagittal
image. C: Gadolinium-enhanced T1-weighted sagittal image revealing enhanced mass lesion at conus medullaris (arrow). D: Anteroposterior view of myelogram.
E: Lateral view of myelogram. Lumbar disk herniation was successfully removed. Myelography showed total blocking of contrast dye at the thoracolumbar
region (D and E).
236 W.-C. Choi et al. / Surgical Neurology 71 (2009) 234–237

Fig. 3. Histologic findings of glioblastoma multiforme. A: The tumor is composed of atypical cells with high cellularity (hematoxylin and eosin stain, original
magnification ×200). B: Moderate cytologic pleomorphism and mitotic aspects. C: Endothelial hyperplasia is prominent. D: Ki-67 is expressed in the nuclei of
tumor cell (N10%) (immunohistochemical stain for Ki-67, original magnification ×200).

reveal any other mass lesion. There was no bony erosion in from the brain, which covers up to 25% of the total
CT scan at the corresponding vertebral body. occurrences. Regardless of its origin, it has a predilection to
develop from the cervical region in the primary cases, and it
2.2. Operation
has a tendency to develop at a young age (b30 years old). In
Surgery consisted of total laminectomy of T11 through 12 comparison with other low-grade gliomas, the progression of
vertebrae with partial removal of the tumor. The tumor mass clinical course is more rapid and dramatic with the time
was gray and friable. It was impossible to distinguish tumor range of several weeks to months. Because of its rapid
margin from spinal cord. Therefore, we only removed the nature, surgery is usually restricted to a role for histologic
mass partially. diagnosis and application of proper adjuvant therapy. With
The pathologic diagnosis confirmed the diagnosis of all these measures, the estimated survival of the patients with
glioblastoma with histologic findings of pleomorphism, intramedullary glioblastoma barely exceeds 12 to 24 months
endothelial hyperplasia, mitotic figure, and palisading [1,5,7-9].
necrosis (Fig. 3). In the current case, the glioblastoma at conus medullaris
masqueraded as L1 through L2 disk herniation initially
2.3. Postoperative course because of the lack of neurologically significant symptoms
After the surgery, the patient was transferred for further and the lack of identification of conus abnormality from
adjuvant therapy. Chemotherapy with temozolomide was initial MRI. When a patient complains of common symptom
prescribed. Radiotherapy was performed as well. of back pain or radiculopathy, every surgeon should pay
On the follow-up 6 months after the second operation, the more attention the routine preenhanced lumbar MRI,
patient still survives; but he became completely paraplegic reviewing not only the degenerative pathology but also the
and totally lost bladder function. potential pathology of the spinal cord. In this current case, at
least 2 months of important therapeutic time was wasted
without doing anything.
Because of its malignant nature, we do not know whether
3. Discussion
timely diagnosis and evacuation of tumor as much as
Intramedullary glioblastoma is a rare disease entity. It can possible would have favorably altered the clinical course of
develop primarily from the cord or as a secondary metastasis this patient. But the lesson that we have learned is that a
 W.-C. Choi et al. / Surgical Neurology 71 (2009) 234–237 237

thorough evaluation from both neurologic and radiologic [4] Claus D, Sieber E, Engelhardt A, et al. Ascending central nervous
examinations should precede any surgical intervention to spreading of a spinal astrocytoma. J Neurooncol 1995;25:245-50.
[5] Cohen AR, Wisoff JH, Allen JC, et al. Malignant astrocytomas of the
avoid missing a coexisting pathology. spinal cord. J Neurosurg 1989;70:50-4.
[6] Cooper PR. Outcome after operative treatment of intramedullary spinal
cord tumors in adults: intermediate and long-term results in 51 patients.
4. Conclusion Neurosurgery 1989;25:855-9.
[7] Grisold W, Pernetzky G, Jellinger K. Giant-cell glioblastoma of the
The neurosurgeon should be suspicious that both the thoracic cord. Acta Neurichir 1981;58:121-6.
intramedullary and degenerative pathologies may coexist in [8] Guidetti B, Mercuri S, Vagnozzi R. Long-term results of surgical
the upper lumbar spine. Careful neurologic and radiologic treatment of 129 intramedullary spinal gliomas. J Neurosurg 1981;54:
evaluations should precede any decision of surgical inter- 323-30.
vention. If combined abnormalities are suspected, a thorough [9] Kopelson G, Linggood RM. Intramedullary spinal cord astrocytoma
versus glioblastoma: the prognostic importance of histologic grade.
radiologic evaluation including a whole-spine MRI with Cancer 1982;50:732-5.
enhancement should be performed. [10] Shirato H, Kamada T, Hida K, et al. The role of radiotherapy in the
management of spinal cord glioma. J Radiat Oncol Biol Phys 1995;33:
Acknowledgments 323-8.

This study was supported by a grant from the Wooridul

Spine Foundation, 272-28, Gwahae-dong, Gangseo-gu,
Seoul, Korea. Commentary

References
Unfortunately, it is very easy to miss a significant lesion
[1] Alvisi C, Cerisoli M, Giulioni M, et al. Intramedullary spinal gliomas: on imaging when the surgeon's initial impression is satisfied
long-term results of surgical treatments. Acta Neurochir 1984;70:169-79. by a secondary lesion. This case history by Won-Chul Choi
[2] Asano N, Kitamura K, Seo Y, et al. Spinal cord glioblastoma and colleagues should remind all neurosurgeons not to fall
multiforme with intracranial dissemination. Neurol Med Chir Tokyo into the same trap as that of the initial surgeon in this case.
1990;30:489-94.
[3] Clappetta P, Salvati M, Cappocia G, et al. A: spinal glioblastoma:
report of seven case and review of the literature. Neurosurgery 1991; Harold D. Portnoy, MD
28:302-6. Pontiac, MI 48341, USA