Intramedullary Spinal Tumors

Intramedullary spinal cord tumors (IMSCTs) arise from within the substance of the spinal cord parenchyma, representing 4-10% of all central nervous system tumors. The most common types are ependymomas (60%), astrocytomas (30%), and hemangioblastomas (3-5%). These tumors present a unique surgical challenge requiring microsurgical techniques to maximize resection while preserving neurological function.

Spinal Cord Microanatomy

Understanding the internal architecture of the spinal cord is essential for surgical planning:

Tumor Histopathology

Pathophysiological Mechanisms of Cord Dysfunction

Syrinx formation - reported in a substantial minority to a majority depending on tumour type; in the hemangioblastoma series cited on this page, peritumoral oedema or syringomyelia was present in 82%:

• Obstruction of CSF flow around tumor
• Dissection of fluid into central canal
• Progressive cavity formation rostral and caudal to tumor
• Syringomyelia symptoms: dissociated sensory loss, weakness
• Syrinx often resolves after tumor resection

Pre-Operative Planning

Microsurgical Resection Technique

Extent of Resection Definitions

Adjuvant Therapy

• A different location and a different grade. Myxopapillary ependymoma arises almost exclusively from the filum terminale / conus medullaris, so it presents as a lumbosacral (cauda equina) lesion - low back pain, radiculopathy, and bladder/bowel or sexual dysfunction - rather than a cervical myelopathy. The 2021 WHO CNS5 classification reclassified it as grade 2 (up from the previous grade 1), formally recognising that it recurs more readily than a cranial grade-1 ependymoma.
• The defining problem is seeding, so remove it en bloc. Because it sits in the CSF-filled thecal sac and has a delicate capsule, violating the capsule risks CSF (drop) metastasis and local recurrence - a particular problem in children. The surgical principle is therefore en-bloc gross total resection, removing the involved filum intact; when the tumour is adherent to conus/roots, safe function-preserving subtotal removal is accepted, with adjuvant radiotherapy. Be precise about which distinction carries the risk, because it is total versus partial removal rather than one piece versus several. In Nakamura's 25-case series (PMID 19934795) nine of fifteen total resections were performed piecemeal and then irradiated, and fourteen of the fifteen were recurrence-free at a mean of 10.4 years, whereas all six patients left with a partial resection and a local radiotherapy field alone died of CSF dissemination.
• Stage the whole neuraxis and consider radiotherapy. Because of the dissemination risk, obtain craniospinal (whole-neuraxis) gadolinium MRI to look for drop metastases, and reserve adjuvant radiotherapy for subtotally resected, recurrent, or disseminated disease. Lifelong surveillance is needed given late recurrence.

• Why the spine deforms after laminectomy. A multilevel laminectomy removes the posterior tension band (spinous processes, laminae, interspinous/supraspinous ligaments and, if violated, the facets). Without it the flexion moment is unopposed, producing progressive post-laminectomy kyphosis - classically a "swan-neck" cervical kyphosis after cervical/cervicothoracic laminectomy.
• Children are the high-risk group. The immature spine is uniquely vulnerable: ligamentous laxity, more horizontally-oriented facets, wedge-shaped cartilaginous vertebrae and ongoing growth mean multilevel laminectomy in a child carries a high rate of progressive kyphosis/scoliosis. Other risk factors are cervical/cervicothoracic level, the number of levels, facet-joint violation (more than 50% resection), and pre-existing malalignment.
• Prevention is a surgical decision made at the time of tumour surgery. Options, in escalating order: preserve the facets (the topic's less-than-50-percent rule) and posterior ligaments as far as the exposure allows; perform an osteoplastic laminoplasty (replace/reconstruct the posterior arch) rather than a laminectomy, especially in children and over multiple levels; and add prophylactic posterior instrumented fusion where the risk is high (extensive facet removal, pre-existing kyphosis, very young child, long-segment cervicothoracic exposure). Established, symptomatic post-laminectomy kyphosis is much harder to treat, usually needing combined anterior-posterior reconstruction and instrumented fusion, so it is far better prevented than corrected.

Global Epidemiology

• Intramedullary spinal cord tumours are rare, comprising roughly 4-10% of all CNS tumours and 20-30% of all intraspinal tumours in adults; they are proportionally more common in children, in whom astrocytoma predominates over ependymoma.
• In adults the order of frequency is ependymoma (around 60%), astrocytoma (around 30%) and hemangioblastoma (3-5%); in the paediatric population astrocytoma is the most common intramedullary tumour.
• Hemangioblastoma is associated with von Hippel-Lindau disease in roughly a quarter of cases; VHL is autosomal dominant (chromosome 3p25, VHL gene) with a 50% risk to offspring.
• Spinal ependymoma shows a slight male predominance and peaks in the third-to-fifth decades; myxopapillary ependymoma characteristically arises at the conus/filum.

Side-by-Side Guidance

There is no single dedicated international guideline for intramedullary tumours; practice is informed by neuro-oncology bodies and surgical society consensus. The table summarises where authoritative sources converge and where emphasis differs.

Registry & Outcome Notes

• Because these tumours are rare, robust data come from population cancer registries (e.g. CBTRUS in the US) and large single-/multi-centre surgical series rather than implant-style joint registries. CBTRUS-type data confirm the rarity and the histological distribution above.
• Reported gross total resection rates from high-volume centres are around 70-90% for ependymoma and over 90% for hemangioblastoma, but under 20-45% for infiltrative astrocytoma - figures that should temper consent discussions.

High- vs Limited-Resource Practice Variation

• High-resource settings: routine whole-spine 1.5-3T MRI with contrast, intraoperative neuromonitoring including D-wave, ultrasonic aspiration, neuro-anaesthesia and ICU support, molecular pathology, and access to conformal radiotherapy and genetic services for VHL.
• Limited-resource settings: MRI access and intraoperative monitoring may be restricted, shifting practice toward later presentation, more conservative (often subtotal) resection to avoid deficit without monitoring, and limited molecular/genetic testing. Outcomes are correspondingly more dependent on pre-operative neurological status, reinforcing the universal principle of operating before deficits become fixed.

Several management questions in intramedullary tumour surgery remain unresolved and are favourite examiner territory because they expose understanding of evidence rather than recall.

• Timing of surgery for the minimally symptomatic tumour. Pre-operative neurological status is the most consistent predictor of outcome, arguing for early resection before deficits become fixed. Against this is the real operative risk of converting a mildly symptomatic patient to a worse state. There is no randomised evidence; practice is individualised to tumour type, growth and patient preference.

• MEP versus D-wave thresholds. A transient MEP drop with a preserved D-wave usually predicts only transient weakness, whereas a D-wave fall over 50% predicts permanent loss. The exact alarm criteria and how aggressively to continue after a transient MEP change are not standardised, and D-wave recording is not feasible in every cord segment or in young children.

• Role of adjuvant radiotherapy for subtotally resected ependymoma. Some series support radiotherapy for residual or anaplastic disease while others, including the original McCormick experience, achieved durable control with surgery alone. There is no level I evidence, and radiotherapy carries risks of myelopathy and second malignancy.

• Radiotherapy for low-grade astrocytoma. Because degree of resection does not change survival within grade and benefit of adjuvant radiotherapy is unproven for grade II tumours, many centres observe after maximal safe resection and reserve radiotherapy for progression - but practice varies widely.

• Surgery versus observation in VHL hemangioblastoma. Asymptomatic lesions can be observed given their indolent, multifocal nature, yet some advocate earlier resection of accessible dorsal lesions before they enlarge. The decision balances cumulative operative morbidity against the risk of symptomatic progression.

• Prophylactic fusion / laminoplasty. Whether to fuse or perform laminoplasty to prevent post-laminectomy kyphosis - particularly in children and after multilevel exposure - is not settled by trial evidence and is guided by age, sagittal alignment and extent of facet preservation.

• Molecular reclassification. WHO CNS5 (2021) increasingly defines spinal ependymoma and glioma by molecular markers (e.g. MYCN amplification in spinal ependymoma, which carries a poor prognosis). How molecular subtype should modify the surgical aggressiveness and adjuvant strategy is still evolving.

1. McCormick PC, Torres R, Post KD, Stein BM. Intramedullary ependymoma of the spinal cord. J Neurosurg. 1990;72(4):523-532. doi:10.3171/jns.1990.72.4.0523. PMID 2319309.

2. Halvorsen CM, Kolstad F, Hald J, et al. Long-term outcome after resection of intraspinal ependymomas: report of 86 consecutive cases. Neurosurgery. 2010;67(6):1622-1631. doi:10.1227/NEU.0b013e3181f96d41. PMID 21107192.

3. Lonser RR, Weil RJ, Wanebo JE, DeVroom HL, Oldfield EH. Surgical management of spinal cord hemangioblastomas in patients with von Hippel-Lindau disease. J Neurosurg. 2003;98(1):106-116. doi:10.3171/jns.2003.98.1.0106. PMID 12546358.

4. Cannizzaro D, Mancarella C, Nasi D, et al. Intramedullary spinal cord tumors: the value of intraoperative neurophysiological monitoring in a series of 57 cases. J Neurosurg Sci. 2022;66(5):447-455. doi:10.23736/S0390-5616.19.04758-1. PMID 31565906.

5. Siller S, Szelenyi A, Herlitz L, Tonn JC, Zausinger S. Spinal cord hemangioblastomas: significance of intraoperative neurophysiological monitoring for resection and long-term outcome. J Neurosurg Spine. 2017;26(4):483-493. doi:10.3171/2016.8.SPINE16595. PMID 27982764.

6. McGirt MJ, Goldstein IM, Chaichana KL, et al. Extent of surgical resection of malignant astrocytomas of the spinal cord: outcome analysis of 35 patients. Neurosurgery. 2008;63(1):55-60. doi:10.1227/01.NEU.0000335070.37943.09. PMID 18728568.

7. Innocenzi G, Salvati M, Cervoni L, Delfini R, Cantore G. Prognostic factors in intramedullary astrocytomas. Clin Neurol Neurosurg. 1997;99(1):1-5. doi:10.1016/s0303-8467(96)00555-0. PMID 9107459.

8. Samartzis D, Gillis CC, Shih P, O'Toole JE, Fessler RG. Intramedullary spinal cord tumors: part I--epidemiology, pathophysiology, and diagnosis. Global Spine J. 2015;5(5):425-435. doi:10.1055/s-0035-1549029. PMID 26430598.

9. Samartzis D, Gillis CC, Shih P, O'Toole JE, Fessler RG. Intramedullary spinal cord tumors: part II--management options and outcomes. Global Spine J. 2016;6(2):176-185. PMID 26933620.

Every card below carries its PMID and DOI.