Acta Neurochir (2010) 152:1139–1144 DOI 10.1007/s00701-010-0620-y
CLINICAL ARTICLE
Non-dysraphic intradural spinal cord lipoma: case series, literature review and guidelines for management Syed Mohammed Rezaul Kabir & Dominic Thompson & Kia Rezajooi & Adrian T. H. Casey
Received: 18 January 2010 / Accepted: 18 February 2010 / Published online: 10 March 2010 # Springer-Verlag 2010
Abstract Purpose Non-dysraphic intradural spinal cord lipomas are rare lesions and the management remains controversial. We present our experience with five cases and propose guidelines for their management. Methods Five patients who underwent surgery for nondysraphic spinal cord lipomas between January 2004 and April 2009 were retrospectively reviewed. All had varying degrees of neurological symptoms at the time of surgery with characteristic features on magnetic resonance imaging (MRI). All patients underwent decompression with a laminectomy/laminoplasty and debulking. The dura was primarily closed in one patient. The literature was also extensively reviewed regarding these rare lesions and optimum management guidelines proposed. Results The age at presentation ranged from 17 to 52 years (mean 32.2). Minimum follow-up was 8 months and maximum follow-up was 5 years. There was neurological improvement following surgery in all cases. Post-operative MRI scan showed evidence of significant residual tumour in all patients. Conclusion The extent of surgical resection does not necessarily correlate with clinical outcome. The aim of S. M. R. Kabir (*) : K. Rezajooi : A. T. H. Casey The National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK e-mail:
[email protected] S. M. R. Kabir : K. Rezajooi : A. T. H. Casey The Royal National Orthopaedic Hospital, London, UK D. Thompson Great Ormond Street Hospital for Children, London, UK
surgery should, therefore, be adequate decompression with preservation of neural structures. Aggressive debulking should be avoided. Onset of any neurological symptoms/ signs, bowel or bladder symptoms or intractable local symptoms should be an indication for surgery. Keywords Intradural lipoma . Non-dysraphic lipoma . Spinal lipoma
Introduction Spinal cord lipomas are rare lesions and comprise only 1% of all spinal cord tumours [1–5]. They are mostly associated with spinal dysraphism [4–7] where these lesions communicate with subcutaneous lipomas through a defect in the posterior elements of the spine. Truly non-dysraphic spinal cord lipomas are very rare [2, 4–13]. Majority of the reported cases are in the pre-MRI era where the diagnosis of occult spinal dysraphism was difficult. Optimum management of these lesions remains controversial. We present our experience with five non-dysraphic cases. We also review the literature and propose guidelines for their management.
Case series Between January 2004 and April 2009, we operated on five cases of non-dysraphic intradural spinal cord lipomas. The summary is in Table 1. There were three female and two male patients. The age at presentation ranged from 17 to 52 years (mean 32.2). Minimum follow-up was 8 months and maximum follow-up was 5 years. In our series, three patients presented with a short history of worsening
1140
S.M.R. Kabir et al.
Table 1 Summary of the patients who underwent surgery for non-dysraphic intradural lipoma Case no.
Age/sex
Symptoms/signs
Location
Surgery
Post-operative outcome
1.
49F
1/12 right buttock/right thigh pain, right leg not feeling normal
T11/Conus
Laminoplasty
Mild numbness right leg— improving No motor deficit MRI: significant residual
2.
3.
26F
17F
4.
52M
5.
17M
No motor weakness, altered sensation right leg, mild hyperreflexia Initially presented at age 15 with 2-year history of neck pain and pain in shoulders, has been managed with regular follow-up; neck pain got worse with worsening balance and pins/needles left hand No motor deficit but very brisk reflexes with sustained clonus 3/12 low back pain, left leg weakness and right leg paraesthesia Increased tone, severe left leg weakness, brisk reflexes with upgoing plantars 3/12 altered sensation/weakness left leg Mild left leg weakness and decreased sensation on examination; brisk reflexes 3 year history of back pain, occasional urinary incontinence and occasional faecal soiling No neurological deficit
Debulking Duroplasty C5-T2
Laminoplasty Debulking
No motor deficit; neck pain much better MRI: significant residual
Primary dural closure T2-T6
T4-T6
Conus
Laminoplasty Debulking Duroplasty
Leg weakness completely recovered No motor/sensory deficit MRI: significant residual
Laminoplasty Debulking Duroplasty Laminoplasty Debulking
Neurologicaly better Very mild weakness left leg MRI: significant residual Urinary symptoms much better No faecal soiling
Duroplasty
MRI: significant residual
F female, M male
neurological symptoms. One patient had a long history of neck pain and was being managed by serial clinical and radiological follow-up. However, she had to be operated because her neck pain became intractable and she developed neurological symptoms and overt upper motor neuron signs. One other patient had a 3-year history of bowel and bladder symptoms. However, there was delay in the initial referral of this patient to us. All patients had pre-operative magnetic resonance imaging (MRI) and the features were virtually diagnostic of spinal cord lipoma (Fig. 1a, b). Two of the tumours were in the conus medullaris, two were entirely thoracic and one was cervicothoracic. The aim of surgery was achieving decompression and preserving neurological function. Somatosensory evoked potential was used in all patients. In addition, motor evoked potential was used in three cases. Three patients underwent laminoplasty and two had laminectomy. In all cases, the lesions presented to the surface and was covered by a thin layer of pia (subpial).They were also found to distort and expand the medulla (juxtamedullary). The debulking was done using standard microneurosurgical principles. The ultrasonic aspirator was used to help with the debulking. There was no clear plane identifiable between the lipoma and spinal cord. The tumour was debulked until it was felt
that adequate decompression has been achieved (Fig. 2). The deliberate surgical strategy was not to attempt to develop any interface between the spinal cord and the tumour. Primary dural closure was achieved in one patient and in the other cases a duroplasty was done with the help of artificial dural substitutes. There were no immediate post-operative complications. All patients improved neurologically following surgery. Post-operative MRI scan showed significant residual tumour in all cases (Fig. 1c, d). The histology was diagnostic of lipoma. All patients underwent regular clinical and radiological follow-up. None of the patients deteriorated or developed any new symptoms. Serial MRI scan did not show any significant change in the residual lesion.
Discussion Non-dysraphic spinal cord lipomas are very rare lesions. Majority of the cases reported were in the pre-MRI era where the diagnosis of occult spinal dysraphism was difficult. In contrast to the dysraphic variety, non-dysraphic lipomas are completely intradural and have no subcutaneous component [14, 15]. They may manifest with local symptoms, e.g., pain or with features of mass effect once they are large enough
Non-dysraphic intradural spinal cord lipoma
1141
Fig. 1 a, b Pre-operative and c, d post-operative sagittal and axial MRI scan of the patient with cervical intradural lipoma showing characteristic high signal on T1-weighted images. Note the presence of significant residual tumour on the post-operative scans
[16]. They tend to be dorsally located in the spinal cord and may demonstrate an exophytic component [14]. True intramedullary lipoma is even rarer [2, 5, 6, 8, 9, 17–22] with few case series reported in the literature.
the inhibition fails and mesenchymal cells form adipocytes [9, 26]. Although all of these theories share some basic aspects, they do not fully explain the exact genesis of the spinal lipoma. However, the first hypothesis is most commonly accepted [9].
Origin Location There are several theories regarding the origin of intradural lipoma. The “developmental error theory” postulates that the lipoma develops as a result of inclusion of misplaced adipocytes during the formation of the neural tube [5, 8, 12, 23–25]. It is therefore not a true neoplasm but a hamartoma or a malformation [23]. This theory can explain why the lipoma is dorsally located and may also explain spinal lipoma without dysraphism [5, 8, 9, 12, 23–25]. There can be peripheral nerve twig, dermoid cyst, skeletal muscles and lymphoid or renal tissue within the fat. These originate from ectoderm or mesoderm [13, 23]. The “metaplasia theory” states that connective tissue metaplasia may lead to deposition of fat within the dura [23, 24]. There is another hypothesis that postulates that adipocytes could arise from cells that give rise to the spinal vessels. Under normal circumstances, mesenchymal cells form the spinal vessel and they are prevented from forming adipocyte by neural crest cells. However, if there are defective neural crest cells,
Non-dysraphic intradural lipomas are commonly found in the thoracic spine, followed by the cervico-dorsal region with only 12% in the cervical cord alone [4, 27, 28]. The characteristic location is in the posterior aspect of the cord within a small radius around the midline. They tend to be present to the surface (subpial) and distort and expand the spinal cord (juxtamedullary). Most involve several spinal cord segments [29]. True intramedullary lipoma is very rare [2, 5, 6, 8, 9, 17–22] with few case series reported in the literature.
Presentation Fifty-five percent of all cases of intradural lipomas present in their second or third decades of life [4]. Around 24% of patients present in their first decade and 16% during the
1142
S.M.R. Kabir et al.
Fig. 2 a–d Intraoperative appearances of the same patient with the cervical lipoma. a, b Appearance following opening the dura. Arrow pointing towards the spinal cord. c Part of the lipoma that has been excised. d Tumour bed following debulking
fifth decade. They are equally prevalent in both sexes [30]. During the first decade of life, spinal lipomas may present as tetraplegia or floppy baby syndrome; the neurological manifestations are thought to be due to birth trauma to the spinal cord [6, 10]. In other patients, however, the manifestation is often delayed [6]. The presentation is often with an ascending spastic paresis affecting one or both legs [4, 27]. Pain when present is usually not radicular but rather localised to the area involved [4, 23, 27]. Most patients are symptomatic for 2 years before seeking medical advice [4, 16]. However, in cases of lipomas restricted solely to the cervical spinal cord, over 80% of patients displayed symptoms for more than 10 years [4]. The symptoms are due to the lesions enlarging as a result of increased fat deposition in metabolically normal fat cells [6].
Radiological appearance MR imaging is the most sensitive modality for diagnosing intradural lipomas preoperatively [6]. Due to the high proportion of fat, there is a short T1 relaxation time. Thus, lipomas have characteristic hyperintense appearance on T1-weighted images (Fig 1). This can be confirmed with fat suppression images. Benign lipomas have relaxation parameters similar to those of subcutaneous fat. Liposarcomas and other fat-related tumours have longer T1 relaxation times and therefore are less intense than subcutaneous fat on T1-weighted images [12, 14].
Treatment Optimum management of these lesions remains controversial and there has been wide variation in their management. Interestingly, this ranges from conservative management with strict diet control to aggressive total removal of the lesion [27]. The fat of the lipoma is metabolically similar to body fat. For this reason, some authors advocate weight loss and strict diet control. Endoh et al. reported a case of an intradural lipoma in a 9-year-old boy that shrank spontaneously with loss of general body fat [31]. Similar observations were also noted by Akyuz et al. [1]. However, this is not always the case as there has been reported cases in which rapid lipoma growth was observed despite scrupulous diet control [32]. The surgical pitfalls in treating these lesions and the dangers of total removal have also long been recognised. Elsberg [33], in 1925, discussed the dangers of total removal of these lesions with the description of a disastrous post-operative quadriplegia. In Ammerman’s series [23], the two patients who underwent laminectomy and total removal developed post-operative paraplegia. The other five patients who had a decompressive laminectomy with either a biopsy or subtotal removal either improved or had minimal deterioration. In the series of six intramedullary lipomas described by Lee and Epstein et al. [5], the patient who underwent 70% resection deteriorated neurologically during the immediate postoperative period. In contrast, the patient who had 40% resection had neurological improvement post-operatively. Both these papers clearly show that the degree of resection
Non-dysraphic intradural spinal cord lipoma
is not directly related to post-operative clinical outcome. As these lesions are indolent in nature and because aggressive resection is associated with significant morbidity, many surgeons advocate doing a bony decompression and wide dural opening and no more than just a biopsy of the lesion [33–35]. It is also advocated that surgery should not be considered in asymptomatic patients [8]. Our cases have some common features with the existing literature. The lesions were mostly thoracic which is the commonest site. All of the cases had classical radiological appearances. However, three of our patients had short duration of symptoms. These symptoms were all neurological in nature and related to the mass effect from the tumour. The main aim of our surgery was achieving decompression with preservation of neurological tissue. Aggressive debulking was not attempted. Primary dural closure without tension was achievable in only one patient. It is noteworthy that all our patients had significant residual tumour in the post-operative scan. However, all patients had improvement in their neurological status postoperatively. They also remained symptomatically stable during the follow-up period. This again shows that the main aim of surgery should be decompression and the degree of resection does not necessarily correlate with the clinical outcome. The appearances of the residual tumour in follow-up serial MRI scans were largely Fig. 3 Proposed guideline for the management of non-dysraphic intradural lipoma
1143
unchanged in all the cases. This is not surprising as intradural lipomas tend to be indolent in nature. Following review of the literature and from our own experience, we propose certain guidelines for the management of these rare lesions (Fig. 3). The main principles of our guidelines are as follows: 1. Asymptomatic patients, patients with local symptoms only, e.g., pain and no neurological signs/symptoms can be managed conservatively with regular clinical and radiological monitoring. 2. In obese patients, weight loss should be advocated. 3. Any patient with worsening or intractable local symptoms or onset of any neurological symptoms or signs or having bowel or bladder symptoms should be considered for surgery. This is because in these patients a point may have been reached when the spinal cord cannot accommodate any more and there is no longer any physiological reserve. Neurological dysfunction then rapidly manifests itself unless decompression is done [5]. 4. The aim of surgery should be decompression with preservation of neurological function. 5. The percentage of tumour debulked does not necessarily correlate with the clinical outcome. Good outcome has been reported with laminectomy and biopsy alone [23]. 6. Aggressive debulking should therefore not be attempted.
1144
7. Duroplasty should be considered if primary dural closure without tension is not possible. 8. Further debulking may be considered if patients continue to deteriorate during follow-up. We hope that these principles will enable a better understanding regarding the overall management of this condition in the future.
References 1. Akyuz M, Goksu E, Tuncer R (2005) Spontaneous decrease in the size of a residual thoracic intradural lipoma. Br J Neurosurg 19:53–55 2. Caram PC, Carton CA, Scarcella G (1957) Intradural lipomas of the spinal cord; with particular emphasis on the intramedullary lipomas. J Neurosurg 14:28–42 3. Ehni G, Love J (1945) Intraspinal lipomas: report of cases; review of the literature, and clinical and pathologic study. J Nerv Ment Dis 102:520 4. Giuffre R (1966) Intradural spinal lipomas. Review of the literature (99 cases) and report of an additional case. Acta Neurochir (Wien) 14:69–95 5. Lee M, Rezai AR, Abbott R, Coelho DH, Epstein FJ (1995) Intramedullary spinal cord lipomas. J Neurosurg 82:394–400 6. Bhatoe HS, Singh P, Chaturvedi A, Sahai K, Dutta V, Sahoo PK (2005) Nondysraphic intramedullary spinal cord lipomas: a review. Neurosurg Focus 18:ECP1 7. Rogers HM, Long DM, Chou SN, French LA (1971) Lipomas of the spinal cord and cauda equina. J Neurosurg 34:349–354 8. Dyck P (1992) Intramedullary lipoma. Diagnosis and treatment. Spine 17:979–981 9. Kim CH, Wang KC, Kim SK, Chung YN, Choi YL, Chi JG, Cho BK (2003) Spinal intramedullary lipoma: report of three cases. Spinal Cord 41:310–315 10. Mori K, Kamimura Y, Uchida Y, Kurisaka M, Eguchi S (1986) Large intramedullary lipoma of the cervical cord and posterior fossa. Case report. J Neurosurg 64:974–976 11. Pani KC, Chopra JS, Kak VK (1972) Intramedullary lipoma of the spinal cord. Neurochirurgia (Stuttg) 15:33–35 12. Patwardhan V, Patanakar T, Patkar D, Armao D, Mukherji SK (2000) MR imaging findings of intramedullary lipomas. AJR Am J Roentgenol 174:1792–1793 13. Timmer FA, van Rooij WJ, Beute GN, Teepen JL (1996) Intramedullary lipoma. Neuroradiology 38:159–160 14. Blount JP, Elton S (2001) Spinal lipomas. Neurosurg Focus 10:e3 15. Wood BP, Harwood-Nash DC, Berger P, Goske M (1985) Intradural spinal lipoma of the cervical cord. AJR Am J Roentgenol 145:174– 176 16. McGillicuddy GT, Shucart W, Kwan ES (1987) Intradural spinal lipomas. Neurosurgery 21:343–346
S.M.R. Kabir et al. 17. Chagla AS, Balasubramaniam S, Goel AH (2008) A massive cervicomedullary intramedullary spinal cord lipoma. J Clin Neurosci 15:817–820 18. Chaskis C, Michotte A, Geffray F, Vangeneugden J, Desprechins B, D'Haens J (1997) Cervical intramedullary lipoma with intracranial extension in an infant. Case illustration. J Neurosurg 87:472 19. El Mostarchid B, Ali A, Maftah M, Mansouri A, Laghzioui J, Kadiri B, Gazzaz M, Chafiq M, Boucetta M (2002) Nondysraphic intramedullary spinal cord lipoma. A case report. Jt Bone Spine 69:511–514 20. Hoffmann G, Warot P, Galibert P, Meignie S, Laine ME (1960) Intramedullary lipomas of the cervicodorsal region. Rev Neurol (Paris) 103:558–567 21. Muthusubramanian V, Pande A, Vasudevan MC, Ramamurthi R (2008) Concomitant cervical and lumbar intradural intramedullary lipoma. Surg Neurol 69:314–317 22. Qureshi NA, Akram M, Ghaffar A, Bhatti S (2006) Nondysraphic intramedullary spinal cord lipoma. J Coll Physicians Surg Pak 16:298–300 23. Ammerman BJ, Henry JM, De Girolami U, Earle KM (1976) Intradural lipomas of the spinal cord. A clinicopathological correlation. J Neurosurg 44:331–336 24. Matsui H, Kanamori M, Miaki K (1997) Expansive laminoplasty for lumbar intradural lipoma. Int Orthop 21:185–187 25. Razack N, Jimenez OF, Aldana P, Ragheb J (1998) Intramedullary holocord lipoma in an athlete: case report. Neurosurgery 42:394– 396, discussion 396–397 26. Catala M (1997) Embryogenesis. Why do we need a new explanation for the emergence of spina bifida with lipoma? Childs Nerv Syst 13:336–340 27. Iwatsuki K (2006) Intradural cervical lipoma with parenchymal marginal fibrous tissue: case report. Neurosurgery 59:E208, discussion E208 28. Lantos G, Epstein F, Kory LA (1987) Magnetic resonance imaging of intradural spinal lipoma. Neurosurgery 20:469–472 29. Drapkin AJ (1974) High cervical intradural lipoma. J Neurosurg 41:699–704 30. Fan CJ, Veerapen RJ, Tan CT (1989) Subdural spinal lipoma with posterior fossa extension. Clin Radiol 40:91–94 31. Endoh M, Iwasaki Y, Koyanagi I, Hida K, Abe H (1998) Spontaneous shrinkage of lumbosacral lipoma in conjunction with a general decrease in body fat: case report. Neurosurgery 43:150–151, discussion 151–152 32. Aoki N (1990) Rapid growth of intraspinal lipoma demonstrated by magnetic resonance imaging. Surg Neurol 34:107–110 33. Elsberg C (ed) (1925) Tumors of the spinal cord and the symptoms of irritation and compression of the spinal cord and nerve roots; pathology, symptomatology, diagnosis, and treatment. Paul B Hoeber New York 34. Foster JJ (1966) Spinal intradural lipomas. A neurosurgical dilemma. Int Surg 46:480–486 35. Swanson H, Barnett J (1962) Intradural lipomas in children. Pediatrics 29:911–926
