Intramedullary tumours are lesions that usually arise directly from the neural tissue of the spinal cord. They are unique amongst all spinal column masses in that they require incision of the spinal cord for surgical access. Intramedullary tumours of the spine make up 2-4% of all central nervous system tumours and are a rare cause of spinal cord dysfunction [1]. Tumours of the spinal column may be classified by location as extradural, intradural, or intramedullary. Amongst all masses of the spinal column, intramedullary tumours are the least common to be encountered in the general public.

Their rare incidence commonly results in misdiagnosis and improper diagnostic workup, resulting in delayed diagnosis. More common clinical entities such as arthritic spinal myelopathy, multiple sclerosis, or even aortic dissection can be confused with intramedullary tumours as they may have similar clinical and radiographic presentations.

Deferred diagnosis and treatment can lead to progressive paralysis, urinary and faecal incontinence, as well as reduced survival. Imaging and surgical technologies were inadequate to diagnose or treat these tumours without serious morbidity in the past. Today, convenient high resolution MRI as well as improved surgical adjuvants and techniques have allowed for significantly improved resection and overall neurological outcome. Most patients can now undergo definitive diagnosis and treatment without significant long term loss of spinal cord function. Thus, awareness of this clinical entity and early diagnosis and treatment is paramount to avoid disability.

Clinical Presentation

Intramedullary spinal cord tumours (IMSCT) have a myriad of presenting signs and symptoms, making a simple diagnositic algorithm difficult. The temporal course of these lesions is widely disparate, with occasional patients presenting with acute neurologic deficits and others with a protracted course. Most series demonstrate that the most common presenting complaint is either local dull pain or radicular pain, often associated with some degree of lower extremity numbness [2,3]. The finding of local back pain when lying flat in bed (nocturnal pain) is highly suggestive of tumour, especially if this pain wakes the patient from sleep; however, these alerting symptoms are often absent. Motor weakness is also common, although it usually presents later than pain or sensory disturbances. The Brown-Sequard syndrome, hemiparesis ipsilateral to the lesion with loss of pain and temperature sensation on the contralateral side, has also been reported in patients with IMSCT. Tumours of the cervical or thoracic spine may lead to lower extremity spasticity. Urinary retention and incontinence are more common in lower cord tumours, although urinary symptoms may be late sequelae of any lesion. Faecal incontinence is much less common. It is important to note that all patients with signs of new or unexplained myelopathy such as spasticity, hyperreflexia, incoordination, or gait disturbance should have further imaging of the spine and appropriate neurological follow-up.

Differential diagnosis

The differential diagnosis of the most common presenting signs and symptoms include IMSCT, intradural extramedullary spinal tumours, epidural spinal tumours, myelopathy due to degenerative disease, cord infarct, vascular lesions such as spinal arteriovenous malformations and dural arterio-venous fistulae, the inflammatory processes such as multiple sclerosis, transverse myelitis and sarcoid. Perhaps most challenging for the general practitioner is differentiating between degenerative spinal canal stenosis, a relatively common condition, and IMSCT (or other spinal tumours), which are relatively rare. The medical history is often helpful, as patients with degenerative disease tend to have years of waxing and waning pain, frequently accompanied by radicular symptoms. Patients with lumbar stenosis often complain of postural back or radicular pain, worst in torso extension (i.e. ascending stairs) and less painful in torso flexion (i.e. pushing a shopping cart). Neurogenic claudication (radicular pain and paraesthesias or numbness with ambulation) due to lumbar spinal stenosis is characteristically relieved by several minutes of sitting, while the radicular pain due to IMSCT is usually not precipitated by walking nor relieved by rest.

Imaging

T1- and T2-weighted MRI with and without gadolinium is the imaging modality of choice for suspected IMSCT. MRI allows the clinician to narrow the broad differential diagnosis listed above, and interpretation by an experienced neuroradiologist is helpful. Ependymomas and astrocytomas have similar imaging characteristics and a definitive diagnosis is only made intraoperatively with a tissue sample. Thus, the role of MRI is not to distinguish between different IMSCT types, but to rule out lesions such as infarct, and auto-immune and inflammatory diseases which will not benefit from surgical intervention. Non-IMSCT lesions tend to have little or no cord enlargement or oedema, thus differentiating them from IMSCT. If the patient is relatively stable and the MRI is equivocal, repeat imaging after one month should show decreased oedema and mass effect in acute auto-immune lesions [4].

Sequences in the sagittal and axial planes are most useful in pre-operative planning. Enhancement on T1-weighted gadolinium sequences is common, despite the low-grade nature of most IMSCT. Associated cysts are also common with IMSCT, and may appear similar to tumour on T1 and T2-weighted images but can be differentiated from tumour by their lack of gadolinium enhancement [4]. IMSCT at any level may be associated with a syrinx; these are especially common in IMSCT of the cervical spine. Once an IMSCT is demonstrated on MRI, prompt referral to a neurosurgeon is warranted. Patients in which MRI is contraindicated may benefit from CT myelography, although this modality is much less useful than MRI. Plain films do not have a significant role in the evaluation of suspected IMSCT.

Treatment

Referral to a neurosurgical specialist for treatment and management of intramedullary masses is very important. This is because, with very few exceptions, all newly diagnosed intramedullary masses require total or subtotal resection or biopsy for tissue diagnosis. Additionally, the surgeon may be able to provide adequate decompression of the spinal cord to avert progression of neurological compromise. It is also possible that a complete surgical resection, if possible, may result in the definitive treatment of many of the tumours of the spinal cord known to be pathologically benign.

Surgery for the resection of intramedullary tumours involves the exposure and decompression of the spinal cord, usually through a multilevel laminectomy followed by a midline dorsal dural opening. Localisation of the laminectomy can be performed using spinal needles and spinal imaging (image intensifier) in the operating theatre. Opening of the spinal cord in order to access the spinal tumour is commonly done through a longitudinal, midline incision. This is done to avoid transection of the white matter tracts of the dorsal columns and avoid disturbing motor and cerebellar long tracts found laterally and ventrally in the spinal cord. Intraoperative ultrasound is commonly employed prior to myelotomy to accurately localise the spinal cord lesion and minimise the extent of the incision [5].

Modern surgical techniques and adjuncts employed during surgery have led to improved neurological outcomes and survival from intramedullary tumours. The employment of the operative microscope as well as ultrasonic aspiration devices during the exposure and resection of the tumour have resulted in an overall neurological morbidity rate of 34% for this type of surgery in current cohorts with most patients improving within 1 month of surgery [6]. Active neuromonitoring during the surgical case has also been a technical advance employed during resection. Combined use of somatosensory evoked potentials (SSEPs) and motor evoked potentials (MEPs) has been shown to reduce surgical morbidity by providing feedback to the surgeon when resection manoeuvres or effects of anaesthesia during the surgery are putting the spinal cord viability at risk [7].

Currently, most neurosurgeons tailor their resections depending on visualised anatomic planes, neuromonitoring changes, and results of intraoperative histology. The presence of a visualised plane between the tumour tissue and the tissue of the spinal cord greatly improves overall resection, neurological outcome, and overall survival [6]. The presence of a syrinx that is continuous with the tumour plane also improves overall neurological outcome [6,8]. The loss of greater than 50% of MEP signals during the case is sensitive for the appearance of significant motor deficits post-operatively.7,9 Another neuromontoring adjuvant, D-wave monitoring, is often employed during surgery and has been shown to correlate with neurological outcome after surgery when used to tailor resections [7].

Non-surgical treatments for intramedullary tumours are largely relegated to patients with diffuse inoperable tumours, those with known incomplete resection, recurrent tumours, or those who could not tolerate surgery or have such a poor prognosis from their primary disease process that surgery would be an ineffective intervention. Historically, external beam radiation has been employed as a treatment for these patients. There exists favourable evidence that radiation likely increases the progression free survival of patients with low grade astrocytomas and ependymomas after partial resection [10,11]. However, evidence is limited on the overall effectiveness of radiation for patients with new progression or malignant pathologies. Although there is an expectation that chemotherapeutic agents may be effective in the treatment of malignant astrocytomas of the spinal cord, there is no evidence that they improve overall survival.

Outcomes

Neurological outcome from surgery is highly correlated with preoperative deficits. Patients with fewer preoperative symptoms are more likely to have good postoperative neurological outcomes [6,12]. Factors predictive of worse postoperative neurological outcomes include thoracic location, advanced age and the presence of urinary symptoms [6,12]. Up to one third of all patients have a significant, acute deterioration in neurological function after surgery, with nearly half returning to their preoperative status within a month of surgery [6].

The most important predictor for tumour recurrence and survival is pathology. Malignant astrocytomas of the spinal cord have an overall recurrence rate of greater than 95%, with outcome unaffected by extent of surgery [13,14]. Complete resection of ependymomas and haemangioblastomas, however, carry a very favourable outcome, with recurrence rates less than 10% over a ten year period [15,16]. Extent of resection does not necessarily correlate with progression-free survival in low grade astrocytomas, however, a growing body of evidence suggests that an increased extent of resection is beneficial to overall neurological outcome in this group of patients [17]. Surgeons are commonly less aggressive with resections if intraoperative histology during the case indicates a diagnosis of astrocytoma and are conversely more aggressive for ependymomas and hemangioblastomas.

Given the greatly improved survival of many of the intramedullary tumour patients, the addition of spinal fusion techniques with internal rod or plate fixation has also been investigated to prevent long-term spinal deformity from the surgery. The presence of preoperative scoliosis, syrinx, long-standing neurological deficit, or cervicothoracic junction location has been correlated with the development of post-operative spinal deformities that can be functionally limiting [18]. Thus the use of osteoplastic laminoplasty or preemptive internal spinal fixation with fusion has been favourably applied with improved prevention of postoperative spinal deformities in children [19].

Currently, outcome-modifying treatments for malignant astrocytomas of the spinal cord do not exist. Patients are empirically treated with postoperative radiation with or without chemotherapy with universally poor results. Aggressive resection, including cordotomy, has not yielded any benefit to outcome and most die of complications of paralysis or progression of disease.

Conclusion


Intramedullary spinal cord tumours are rare but important clinical entities that now can be easily diagnosed and often effectively treated. Importance should be given to recognising clinical and radiographic findings that are associated with these tumours. Surgical advances have now made many of these tumours treatable with acceptable long-term neurological outcomes.

References

  1. Kane PJ, el-Mahdy W, Singh A, Powell MP, Crockard, HA. Spinal intradural tumours: Part II–Intramedullary. Br J
    Neurosurg. 1999:13(6);558-63.https://doi.org/10.1038/sj/leu/2401364
  2. Houten JK and Cooper PR. Spinal cord astrocytomas: presentation, management and outcome. J Neurooncol.
    2000;47(3):219-24.
  3. Schwartz TH and McCormick PC. Intramedullary ependymomas: clinical presentation, surgical treatment strategies and prognosis. J Neurooncol, 2000;47(3):211-8. https://doi.org/10.1023/A:1006414405305
  4. Waldron JS and Cha S. Radiographic features of intramedullary spinal cord tumors. Neurosurg Clin N Am.
    2006;17(1):13-9.
  5. Epstein FJ, Farmer JP and Schneider SJ. Intraoperative ultrasonography: an important surgical adjunct for intramedullary tumors. J Neurosurg, 1991;74(5):729-33. https://doi.org/10.3171/jns.1991.74.5.0729
  6. Garces-Ambrossi GL, McGirt MJ, Mehta VA et al. Factors associated with progression-free survival and long-term neurological outcome after resection of intramedullary spinal cord tumors: analysis of 101 consecutive cases. J Neurosurg Spine. 2009;11(5):591-9.
  7. Sala F, Palandri G, Basso E, et al. Motor evoked potential monitoring improves outcome after surgery for intramedullary spinal cord tumors: a historical control study. Neurosurgery. 2006;58(6):1129-43; discussion 1129-43.
  8. Samii M and Klekamp J. Surgical results of 100 intramedullary tumors in relation to accompanying
    syringomyelia. Neurosurgery, 1994;35(5):865-73; discussion 873. https://doi.org/10.1097/00006123-199411000-00010
  9. Quinones-Hinojosa A, Lyon R, Zada G, et al. Changes in transcranial motor evoked potentials during intramedullary spinal cord tumor resection correlate with postoperative motor function. Neurosurgery. 2005;56(5):982-93; discussion 982-93. https://doi.org/10.1007/s00105-005-1035-1
  10. Chun HC, Schmidt-Ullrich RK, Wolfson A, Tercilla OF, Sagerman RH, King GA. External beam radiotherapy for primary spinal cord tumors. J Neurooncol, 1990;9(3):211-7. https://doi.org/10.1007/BF02341151
  11. Linstadt DE, Wara WM, Leibel SA, Gutin PH, Wilson CB, Sheline GE. Postoperative radiotherapy of primary spinal
    cord tumors. Int J Radiat Oncol Biol Phys. 1989;16(6):1397-403.
  12. McGirt MJ, Chaichana KL, Atiba A, Attenello F, Yao KC, Jallo GI. Resection of intramedullary spinal cord tumors in children: assessment of long-term motor and sensory deficits. J Neurosurg Pediatr. 2008;1(1):63-7.
  13. Cohen AR, Wisoff JH, Allen JC, Epstein F. Malignant astrocytomas of the spinal cord. J Neurosurg, 1989;70(1):50-4. https://doi.org/10.3171/jns.1989.70.1.0050
  14. Constantini S, Miller DC, Allen JC, Rorke LB, Freed D, Epstein FJ. Radical excision of intramedullary spinal cord tumors: surgical morbidity and long-term follow-up evaluation in 164 children and young adults. J Neurosurg.
    2000;93(2 Suppl):183-93.
  15. Binning M, Klimo P Jr, Gluf W, Goumnerova L. Spinal tumors in children. Neurosurg Clin N Am. 2007;18(4):631-58.
  16. Epstein FJ. Spinal cord tumors in children. J Neurosurg. 1995;82(3):516-7.
  17. Jallo GI, Danish S, Velasquez L, Epstein F. Intramedullary low-grade astrocytomas: long-term outcome following radical surgery. J Neurooncol, 2001;53(1):61-6. https://doi.org/10.3406/mat.2001.403256
  18. McGirt MJ, Chaichana KL, Atiba A, Bydon A, Witham TF, Yao KC, Jallo GI. Incidence of spinal deformity after resection of intramedullary spinal cord tumors in children who underwent laminectomy compared with laminoplasty. J Neurosurg Pediatr. 2008;1(1):57-62.
  19. Simon SL, et al. Efficacy of spinal instrumentation and fusion in the prevention of postlaminectomy spinal deformity in children with intramedullary spinal cord tumors. J Pediatr Orthop. 2008;28(2):244-9. https://doi.org/10.1097/01.wno.0000312735.05048.fb