Cystic spinal dysraphism of the cervical and upper thoracic region

Transcription

1 Childs Nerv Syst (2006) 22: DOI /s ORIGINAL PAPER J. Francisco Salomão Sérgio Cavalheiro Hamilton Matushita René D. Leibinger Antonio R. Bellas Elide Vanazzi Luiz A. M. de Souza Andréa G. Nardi Cystic spinal dysraphism of the cervical and upper thoracic region Received: 28 December 2004 Published online: 4 June 2005 # Springer-Verlag 2005 J. F. Salomão (*). R. D. Leibinger. A. R. Bellas Section of Pediatric Neurosurgery, Department of Pediatric Surgery, Fernandes Figueira Institute, Oswaldo Cruz Foundation (M.S. Fiocruz), Av. Rui Barbosa, 716, Rio de Janeiro, Brazil fsalomao@ism.com.br Tel.: Fax: S. Cavalheiro Section of Pediatric Neurosurgery, Federal University of São Paulo, Rua Botucatu 591/42, São Paulo, Brazil H. Matushita Section of Pediatric Neurosurgery, São Paulo University Medical School, Av. Dr. Enéas de Carvalho Aguiar, 255, São Paulo, Brazil E. Vanazzi Department of Pathology, Fernandes Figueira Institute, Oswaldo Cruz Foundation (M.S. Fiocruz), Av. Rui Barbosa, 716, Rio de Janeiro, Brazil L. A. M. de Souza. A. G. Nardi CT Scan Centro de Diagnóstico, Rua Santo Amaro, 80, Rio de Janeiro, Brazil Abstract Background: Cystic dysraphic lesions of the cervical and upper thoracic region are rare and only a few series have been published about the topic. These malformations can be divided into categories that include both myelocystoceles and the so-called cervical meningoceles or myelomeningoceles. Methods: A retrospective study of 18 patients was conducted. Results: In 17 patients a squamous or a cicatricial epithelium of variable thickness covered the dome of the lesions, while the base was covered with full-thickness skin. In one case the skin was entirely normal. Four patients displayed associated CNS malformations and three more had systemic congenital anomalies. All patients underwent surgical exploration and the length of time between birth and surgery ranged from 6 h to 9 months. The most frequent surgical finding, seen in 14 patients, was a stalk connecting the dorsal surface of the spinal cord to the cyst. In three patients the findings were consistent with myelocystocele. Only in one case was a true meningocele found. Hydrocephalus and Chiari II malformation were not as consistently associated as in myelomeningoceles. Neurological signs and symptoms were not so marked as in myelomeningoceles and were found in the follow-up of four patients. In two of them there was a non-progressive deficit, probably expressing an imperceptible involvement of the nervous system in the first year of life. The histopathological findings were of three types: neuroglial stalks, fibrovascular stalks and myelocystoceles. Conclusions: Cystic dysraphisms of the cervical and upper thoracic region differ clinically and structurally from meningomyelocele and have a more favorable outcome. We believe that these malformations have not been properly labeled and propose a classification based on the structures found inside the cyst. Keywords Spinal dysraphism. Myelomeningocele. Cervical myelomeningocele. Meningocele. Myelocystocele. Spina bifida cystica

2 235 Introduction Spina bifida cystica (SBC) preferentially affects the lower regions of the spine and is rarely seen in the cervical and upper thoracic spine. The posterior midline cutaneous mass lesions of this region have been described under different labels: cervical meningoceles [7, 36], cervical myelomeningoceles [26], cervical spine atretic myelomeningocele [9], rudimentary meningocele [10], myelocystocele [4, 7, 23, 36, 39]. Unlike low thoracic and lumbosacral myelomeningoceles, these malformations are epithelized and the neurological impairment is usually discreet or even absent. In the most common form, a stalk of aberrant nervous tissue tents the dorsal surface of the spinal cord to the posterior dura mater and, through a posterior spina bifida and myofascial defect, to the cystic walls [7, 26, 36]. In some cases, the stalk is composed exclusively of fibrovascular elements [37, 38] that also tether the spinal cord to the dura mater and other soft tissues. The less frequent forms include myelocystoceles [4, 7, 23, 36, 39] and leptomeningeal-lined cysts filled with cerebrospinal fluid (CSF), without any structureinsidethesac[1, 7, 11, 24, 25]. Our purpose is to report a series of this peculiar form of spinal dysraphism, emphasizing their differences from classic myelomeningoceles. Additionally, it is our aim to propose a classification of these lesions, considering the structures found inside the cyst. Patients and methods Our series consists of 18 patients with cystic spinal dysraphism of the cervical and upper thoracic region (CDCT), who were treated at the pediatric neurosurgery divisions of Fernandes Figueira Institute, Oswaldo Cruz Foundation (M. S. Fiocruz), Federal University of São Paulo (EPM UNI- FESP) and São Paulo University Medical School (HC USP). All cases were retrospectively studied and all available data regarding age, gender, timing of surgical procedures, hydrocephalus, Chiari II malformation (CM II) and hydromyelia were recorded as well as associated malformation. The pre- and postoperative images and the pathological specimens available were reviewed. Immediate and late results were recorded and, when needed, patients were called on for new neurological evaluation or neuroimaging studies. A summary of the findings is shown in Table 1. Results Pre- and perinatal features Ultrasonographic prenatal diagnosis was established in eight patients. In two of them, MRI was also obtained (Fig. 1). Eleven patients were females. The largest diameter of lesions ranged from 2 to 16 cm. All the masses were located in the midline, but one extended laterally. In all patients but one, the dome was covered either with a squamous epithelium or with a fibrous tissue of variable thickness (Fig. 2). In all cases the base and most of the wall was covered by full-thickness skin. In one single patient, a CSF leak was observed. Ten lesions were in the cervical spine. In 16 patients, the lesion exited the spinal canal through a single posterior arch defect. In one case, two levels were affected, and in another one, five levels. One patient had a marked respiratory distress since birth and, in the others, the clinical examination did not reveal abnormalities of any kind. Seven newborns had other congenital malformations. Preoperative images were obtained in ten patients and included ultrasonography, CT scan and MRI scan. MRI was obtained in eight patients, providing details from the cele content and other malformations such as hydromyelia, hydrocephalus and CH II (Fig. 3). Surgery All patients had been operated on and the length of time between birth and the first surgical procedure ranged from 6 h to 9 months. In one patient the lesion was tied off with a ligature and the sac removed in one piece, without any further exploration. In all other patients, a two (or more)- level laminectomy or laminotomy was performed and, when identified, the stalks were excised flush to the spinal cord, whenever possible. The surgical findings were of three kinds: (1) Stalks. In 14 patients, stalks of a more or less vascularized tissue emanated from the dorsal surface of the spinal cord, penetrating the sac through a defect of the posterior midline structures and adhering to the cystic wall, eccentrically or at its dome (Fig. 4a). In two patients, the stalks ended in nodules at the base of the cyst (Fig. 4b). In one of these, the cord had a longitudinal split distal to the stalk s base, with the two halves separated by a fibrous septum, characterizing the type II split cord malformation (SCM II). (2) Myelocystoceles. Found in three patients. In all of these there was a second cyst herniating into a meningocele (Fig. 4c). The outer cyst (meningocele) was in continuity with the subarachnoid space and the most internal (myelocystocele), in connection with a hydromyelic central canal. (3) Meningocele. In one patient, a herniation of the meninges filled with CSF was found. Some arachnoid bands tethered the normal spinal cord dorsally and no neural elements were found inside the cyst (Fig. 4d).

3 236 Table 1 CDCT: summary of the clinical findings Case/ Sex Age origin 1 IFF F 48 h MCC T4 T5 2 IFF M 12 days Stalk C2 C3 3 IFF M 48 h Stalk C6 C7 4 IFF M 72 h Stalk T4 T5 5 IFF F 72 h Stalk T1 T2 6 IFF F 9 days MCC C3 T1 7 IFF F 48 h Stalk C4 C5 8 IFF M 24 h Stalk T3 T4 9 IFF M 96 h Stalk C2 C3 10 F 8 d Stalk C3 IFF C4 11 F 2 MC T6 IFF months T7 12 F 6 h Stalk T2 EPM T3 13 M 6 h MCC T1 EPM T2 14 F 10 h Stalk T2 EPM T3 15 M 6 Stalk C3 USP months C4 16 F 30 days Stalk C5 USP C6 17 F 9 Stalk C3 USP months C5 18 USP F 5 months Type Level Systemic Nervous Hydrocephalus Chiari Hydromyelia Type Pathology Follow-up malformations system malformations of surgery Stalk C3 C4 Dermoid Yes Yes Yes Lam, MCC Paresis/left upper limb hypotrophy. Craniovertebral decompression Klippel Feil Neuroschisis No No No Lam FVS Uneventful No No No Tying NGS Spasticity Pyramidal signs SCM II No No No Lam NGS Uneventful S. bifida occulta Yes No No Lam, NGS Uneventful Multiple Yes Yes Yes Lam, MCC Death Yes No No Lam, NGS Mental retardation No No No Lam FVS Uneventful No No No Lam NGS Uneventful Yes Yes No Lam NGS Uneventful No No No Lam NA Uneventful Polycystic No No No Lam NGS Uneventful kidney Rib anomalies No No Yes Lam NGS Uneventful No No No Lam NGS Uneventful No Yes Yes Lam NA Uneventful No Yes Yes Lam NA Lower limb spasticity Yes Yes No Lam, NA Mental ETV retardation. Tetraparesis Yes Yes Yes Lam, NA Mental retardation Lam Laminectomy/laminotomy, FVS fibrovascular stalk, SMC II split cord malformation type II, MC meningocele, NGS neuroglial stalk, ETV endoscopic third ventriculostomy, MCC myelocystocele, NA not available Postoperative course One patient with multiple malformations died 32 days after surgery and the causa mortis was septicemia. The remainder had an uneventful postoperative course without any noticeable change in their neurological status. Hydrocephalus, CH II and hydromyelia Every patient underwent at least one MRI examination, either pre- or postoperatively, in order to study the relationships between hydrocephalus, CH II and hydromyelia. CH II was identified in seven patients. Hydrocephalus

4 237 Fig. 1 Prenatal images. a Gestational ultrasonography showing a cystic mass in the upper cervical region. b Sagittal intrauterine MRI showing a high cervical mass contacting the skull was detected in seven patients and all needed to be operated on. In six patients, either pre- or postoperative images revealed hydromyelia, which was invariably located at the level of the dysraphism, extending upwards or downwards. Its relationship with hydrocephalus and CH II can be seen in Table 1. Follow-up Seventeen patients have a regular follow-up ranging from 2 months to 13 years. One patient (case 3) deteriorated with progressive motor impairment. Postoperative MRI showed that the spinal cord was tethered to the dura mater by remnants of the neuroglial stalk. This patient, the only one in the series that underwent a ligature of the lesion at the neck, was reoperated and the cord released from the adherences. Another patient underwent a craniovertebral decompression due to symptomatic hydromyelia. Two other patients presented some neurological abnormalities that did not progress until now. In three patients some degree of mental retardation was observed. Postoperative MRI studies were performed in 11 patients. In three there were no abnormalities, except for a widening of the subarachnoid space at the operated level. In five, triangular spurs tenting the dorsal aspect of the spinal cord were identified. In these, no progressive neurological deficit was noticed. In one patient the MRI showed a SCM II. Fig. 2 Cystic spinal dysraphisms of the cervical and upper thoracic region (CDCT). The dome of the lesion is covered with squamous epithelium (a) or with a fibrous tissue of variable thickness (b). One patient (c) presented with a huge cyst and the skin was entirely normal

5 238 Fig. 3 Neuroimaging. a CT scan showing, apart from the spina bifida, a stalk penetrating the cyst through the neck of the lesion. b MRI showing the stalk connecting the dorsal surface of the cord to the cyst (arrow). c CH II malformation and hydromyelia can also be seen. d MRI features of a myelocystocele: a cyst inside another cyst is seen in connection with a hydromyelic cavity (arrow) Pathology The pathological specimens available from 13 patients were reviewed. The main histopathological findings were as follows: 1. Neuroglial stalk (eight out of 13 cases): structures composed of dysplastic tissue formed by glia, neurons, ganglia, peripheral nerves, ependima, blood vessels and meningoepithelial cells, sometimes mimicking the structure of an aberrant spinal cord (Fig. 5a and b). 2. Myelocystoceles (three out of 13 cases): cystic structures, whose walls contained aberrant glial and neural tissue, partially lined with ependimary cells (Fig. 5c) 3. Fibrovascular stalks (two out of 13 cases): structures composed of fibroconnective tissue and blood vessels with no neural elements (Fig. 5d). available literature, we conclude that ours is the largest series presented so far. CDCT are seldom reported [2, 3, 6, 8 11, 16, 17, 23 25, 28, 31, 32, 34, 39, 41], and few series have been published [7, 20, 21, 27, 29, 36]. Of these studies, the most extensive one focused on the urologic outcome of 11 patients [29]. The incidence of CDCT varies from 1 to 6.5% of the total cases of SBC [3, 8, 22, 29], and among 296 consecutive cases operated at IFF Fiocruz from 1989 to 2004, we found 11 defects (3.7% of the cases). Such dysraphisms are a heterogeneous group of congenital malformations that share some common characteristics: they are all cystic, have no placode and are located in the upper half of the spinal cord. However, when compared with each other, they show some structural variability (as seen in our series), in which three types of lesions were identified: stalks, myelocystoceles and meningoceles. The data collected from the literature and those reported here lead us to classify these lesions in the following three types, according to the structures found inside the cyst. Discussion We reported on 18 patients having cystic spinal dysraphism with the lower limit at T6 T7. After an extensive review of

6 239 Fig. 4 Surgical findings. a Stalk adhered to the dome of the lesion. b Basal nodes. c Myelocystocele: a second cyst (arrow) is seen inside a meningocele. d Meningocele with no structure inside the cyst. The normal spinal cord, already freed from arachnoidal adhesions, remains confined to the vertebral canal Fig. 5 Histopathological findings. a Structure similar to an aberrant spinal cord with a central canal (H/E 40). b Structures similar to peripheral nerves and ganglia (H/E 100). c Myelocystocele showing an ependimal-lined cavity, neuroglial tissue and a few connective fibers (Gomory s trichromic 400). d Fibrovascular stalk showing only fibroconnective tissue and blood vessels. There are no neural elements (Gomory s trichromic 100)

7 240 (1) Type I or CDCT with stalk, found in 14 out of 18 patients of our series, in which three subtypes were identified: (a) neuroglial stalks, reported under different labels [20, 26, 37]; (b) neuroglial stalks with an underlying SCM II [26]; (c) fibrovascular stalks, also labeled meningoceles by some authors [20, 35 37, 43]. (2) Type II or Myelocystoceles [20, 23, 36, 37, 39]. Found in three patients, in which a second ependymal-lined cyst herniates inside a meningocele. (3) Type III or CDTC without stalk. A rarer form [1, 7, 11, 20, 25], found in only one patient of our series and in which no structures other then CSF, arachnoid bands or free-floating nerve roots were seen inside the cyst. These malformations are, in the strict sense, the true meningocele. CDCT result from defects more limited than those seen in classic myelomeningoceles. According to Pang and Dias [26], the incomplete fusion of the posterior part of the neural tube would give rise to a limited dorsal myeloschisis (LDM). This LDM would differ from typical myelomeningoceles only in the degree of neurulation and a persistent endomesenchimal tract would be the cause of SCM II. This theory does not explain the development of other forms, such as myelocystoceles and true meningoceles. The unifying hypothesis proposed by Steinbok and Cochrane [35, 37] attempts to explain the origin of both meningoceles and myelocystoceles. Its rationale is based on the presence or absence of hydromyelia, its influences on LDM and the effect of hydraulic forces. The authors do not explain the genesis of the SCM II, which was reported in four of the cases described by Pang and Dias [26] and also seen in one of our patients. The location of SBC at different levels is not fortuitous. Neurulation is usually considered a continuous process, starting in the cervical region and progressing rostrally and caudally until it reaches the anterior and posterior neuropores, respectively [15]. Evidences of multiple closure sites have been demonstrated in animals as well as in humans [14, 42]. Van Allen et al. [42] illustrate with a cervical meningocele the incomplete fusion of the rostral part of Closure 1. For these authors, the multisite neural tube closure would be most likely controlled by separate genes, subject to multiple influences. The influence of homeobox genes over the segmental development of the nervous system rules out any possibility that the lesions would be located in the upper spinal cord purely by chance [18]. The initial neurological examination of a newborn with CDCT is usually considered as normal [1, 2, 6, 20, 21, 26, 31, 37, 43] due to the difficulties in recognizing more subtle alterations. It seems that some time is needed to detect discreet anomalies, which would become noticeable only when the use of the limbs is increased [38]. Patients may show a late worsening with motor deficits, pyramidal syndrome, suspended sensory loss, spasticity, proprioceptive disturbances and mirror movements becoming evident [9, 20, 21, 25, 26, 38, 44]. Progressive neurological deficit has been reported in nontreated children and adults [9, 20, 43, 44]. An inadequate surgical technique can lead to late neurological deterioration [11, 21, 26] and therefore, limited procedures should be avoided. An effective approach involves the exposure of the spinal cord for at least two vertebral levels in order to have the lesion properly explored [20, 26, 36]. Stalks, bands, aberrant roots or any other abnormal structures, as soon as they are identified, must be excised flush to the spinal cord [26, 36]. Neurological deterioration is also reported, despite an adequate operation [38]. Three of our patients had impaired intellectual function. Meyer-Heim et al. [21] reported similar findings and believe that it could be related to posterior fossa changes and hindbrain herniation. The incidence of hydrocephalus and CH II in our series concurs with the literature. Unlike classic myelomeningoceles, hydrocephalus can be unrelated to cerebellar herniation. Curiously, the CH II is usually less severe in CDCT. Most of the anomalies associated with CH II seem to be due to leakage of CSF [13, 19]. According to Gardner [13], the pulsations of the anterior choroid plexus are virtually unopposed due to the escape of CSF through a ruptured neural tube. As a consequence, the tentorium will be pushed too far. McLone and Knepper [19] believe that the opening in the neural tube would allow a persistent venting of CSF from the central nervous system during embryonic and fetal life. The defective development of the rhomboencephalic ventricle would influence both the form and the structure of the chondrocranium and brain stem. In both theories the end result is a small posterior fossa, unable to accommodate its structures. It is possible that in CDCT, the less severe CH II results from a less severe leak of CSF. It would also last a shorter period of time, because these defects almost always end up covered by epithelium, which prevents the CSF from leaking. Hydromyelia was found in six of our patients, five of which had CH II. Associated congenital malformations have been described, mostly occult dysraphisms [7, 11, 12, 29], and their influence on late neurological deterioration has been emphasized [7, 11, 26]. Consequently, MRI of the entire spine is recommended [7, 11]. CDCT are the issue of many different denominations [3, 7 10, 26]. Pang and Dias [26] termed myelomeningocele the lesions we labeled CDCT with stalk, due to the presence of nervous tissue inside of a sac lined with meninges. True myelomeningoceles are protrusions of the spinal cord and its membranes through a defect in the vertebral column [33], and their main characteristic is a neural placode, expression of a failure in the closure of the neural tube. The neural placode is, almost invariably, exposed to the environment and there is a variable exten-

8 241 sion of arachnoid in its margins. The placode prevents the fusion of the cutaneous ectoderm from each side, and the same happens with the bony and myofascial tissues. The result is a complex malformation with devastating effects, associated with hydrocephalus and CH II in almost all cases. Myelomeningoceles are found in the lower segments of the thoracic spine and in the lumbar and sacral regions [3, 8, 16, 41]. The neurological deficit is always related to the level of the lesion, which means to say that the higher the lesion, the more marked the deficit. The characteristics of CDCT are distinct from the ones described above. They are more limited and covered by normal skin, except for the dome, lined with squamous epithelium or with a thick scar tissue. There is no neural placode and the morphology of the spinal cord is normal or close to normal. The arachnoid membrane is not exposed and CSF leakage is seldom seen. Hydrocephalus and CH II are less frequently reported. In CDCT, the relationship between neurological deficit and the level of the lesion is less obvious. Consequently, tetraplegia or paraplegia is not part of the clinical features. The neurological function, in a preliminary evaluation, is preserved or slightly impaired, usually at the level of the dorsal columns [21, 26]. Such minor symptoms are related to the discreet bone defects which contrast, in intensity as well as in extension, with those found in myelodysplasias affecting the thoracolumbar region [3]. Most of the lesions termed meningoceles [20, 35 37, 43] are very similar to those labeled myelomeningoceles [2, 26, 31]. The classic definition of meningocele is that of a cystic swelling of the meninges protruding through a spinal defect, with the spinal cord remaining entirely confined to the vertebral canal [5]. It is understood that normal neural components such as nerve roots can be seen floating in the hernia sac or tethered to the neck [8, 11, 30], but not originating from the dorsal surface of the cord, tethering it and ending attached to the cystic walls. Other elements inside the meningocele are seldom mentioned [40] and their presence, in our belief, impairs the definition. Thus, the denomination applied to most cases reported as cervical myelomeningoceles or meningoceles is not correct. True meningocele, in the strictest sense [1, 7, 11, 24, 25], are rarely found. No repair is done to the term myelocystocele that, although histologically distinct from the other forms here described, can be clinically indistinguishable from them. That is the reason why they were included in the series we are currently reporting. In conclusion, the distinction between the various forms of CDCT is often impossible in clinical grounds: the nomenclature is confusing and, in most cases, inadequate. Thus we suggest that while waiting for a most suitable denomination, these lesions should be called generically cystic spinal dysraphisms of the cervical and upper thoracic region (CDCT). The classification based in the structures found inside the cyst seems to be a reasonable alternative. References 1. Ando S, Moritake K, Takaia M (1993) MRI demonstration of cervical meningocele. Neuroradiology 35: Ankola PA, Fernandes Y, Tunnessen WW (1998) Picture of the month. Cervical mielomeningocele. Arch Pediatr Adolesc Med 152: Barson AJ (1970) Spina bifida: the significance of the level and the extent of the defect to the morphogenesis. Dev Med Child Neurol 12: Bhargawa R, Diamond DI, Benzie RJ, Ventureyra ECG, Higgins MJ, Martin DJ (1992) Prenatal demonstration of a cervical myelocistocele. Prenat Diagn 12: Cameron AH (1956) The spinal cord lesion in spina bifida cystica. Lancet 2: Choudhury AR, Gonog MA, Mahmood K (1996) Heterotopic spinal cord? A curiosity. Childs Nerv Syst 12: Delashaw JB, Park TS, Cail WM, Vollmer DG (1987) Cervical meningocele and associated spinal anomalies. Childs Nerv Syst 3: Doran PA, Guthkelch AN (1961) Studies in spina bifida cystica. I. General survey and reassessment of the problem. J Neurol Neurosurg Psychiatry 24: Duprez TP, Laterre EC (1995) Unusual form of closed dysraphysm of the cervical spine. Acta Neurol Belg 95: El Shabrawi-Caelen L, White WL1, Soyer HP, Kim BS, Frieden IJ, McCalmont TH (2001) Rudimentary meningocele: remnant of a neural tube defect? Arch Dermatol 137: Erşahin Y, Barçin E, Mutluer S (2001) Is meningocele really an isolated lesion? Childs Nerv Syst 17: Feltes CH, Fountas KN, Dimopoulos VG, Escurra AI, Boev A, Kapsalaky EZ, Robinson JS, Troup EC (2004) Cervical meningocele in association with spinal abnormalities. Childs Nerv Syst 20: Gardner WJ (1965) Hydrodynamic mechanism of syringomyelia: its relationship to myelocele. J Neurol Neurosurg Psychiatry 28: Golden JA, Chernoff GF (1993) Intermittent pattern of neural tube closure in two strains of mice. Teratology 47: Heuser CH, Streeter GL (1941) Development of the macaque embryo. Contrib Embryo 29: Lorber J (1973) Early results of selective treatment of spina bifida cystica. Br Med J 4: Matson DD (1969) Neurosurgery of infancy and childhood, 2nd edn. Thomas, Springfield, IL 18. McLone DG (1998) The biological resolution of malformations of the central nervous system. Neurosurgery 43: McLone DG, Knepper PA (1989) The cause of Chiari II malformation: a unified theory. Pediatr Neurosci 15: May D, Rilliet B, Berney J (1992) Méningocèles et méningomyélocystocèles cervicales. A propos de 4 cas. Neurochirurgie 38:

9 Meyer-Heim AD, Klein A, Boltshauser E (2003) Cervical myelomeningocele. Follow-up of five patients. Eur J Paediatr Neurol 7: Muñoz BM, Diaz Llopis I, Martinez Angullo E, Lopez A, Garcia Aymerich V, Forner Valero JV (1994) A multicentre study of the hospital care of 1500 patients with mielomeningocele. Paraplegia 32: Nishino A, Shirane R, So K, Arai H, Suzuki H, Sakurai Y (1998) Cervical myelocystocele with Chiari II Malformation: magnetic ressonance imaging and surgical treatment. Surg Neurol 49: Nishio S, Morioka T, Hikino S, Fukui M (2001) Cervical (myelo)meningoncoele: report of 2 cases. J Clin Neurosci 8: Odabasi Z, Gökçil Z, Kütürkçü Y, Vural O, Yardim M (1998) Mirror movements associated with cervical meningocele: case report. Minim Invasive Neurosurg 41: Pang D, Dias MS (1993) Cervical myelomeningoceles. Neurosurgery 33: Pang D, Dias MS, Ahab -Barmada M (1992) Split cord malformation. I. A unified theory of embryogenesis for double spinal cord malformations. Neurosurgery 31: Penfield W, Coburn DF (1938) Arnold Chiari malformation and its operative treatment. Arch Neurol Psychiatry 40: Perez LM, Wilbanks JT, Joseph DB, Oakes WJ (2000) Urological outcome of patients with cervical and upper thoracic mielomeningocele. J Urol 164: Raimondi AJ (1987) Pediatric neurosurgery. Theoretic principles. Art of surgical techniques. Springer, Berlin Heidelberg New York 31. Rosenberg D, Canterino I (1992) Myéloméningocèle cervico-dorsale; syringomyélie. Pédiatrie 47: Smith GK, Smith ED (1973) Selection for treatment in spina bifida cystica. Br Med J 4: Stedman s Medical Dictionary (1995) 26 edn. Williams and Wilkins, Baltimore 34. Stein SC, Schut L, Ames MD (1974) Selection for early treatment in myelomeningocele: a retrospective analysis of various selection procedures. Pediatrics 54: Steinbok P (1995) Dysraphic lesions of the cervical spinal cord. Neurosurg Clin N Am 6: Steinbok P, Cochrane DD (1991) The nature of congenital posterior cervical or cervicothoracic midline cutaneous mass lesions. J Neurosurg 75: Steinbok P, Cochrane DD (1995) Cervical meningoceles and myelocystoceles: a unifying hypothesis. Pediatr Neurosurg 23: Sun JCL, Steinbok P, Cochrane DD (2000) Cervical myelocystoceles and meningoceles: long-term follow-up. Pediatr Neurosurg 33: Suneson A, Kalimo H (1979) Myelocystocele with cerebellar heterotopia. Case report. J Neurosurg 51: Talwalker VC, Dastur DK (1970) Meningoceles and meningomyeloceles (ectopic spinal cord). Clinicopathological basis of a new classification. J Neurol Neurosurg Psychiatry 33: Tomlinson BE (1965) Heterotopic, non-functioning masses of nervous tissue in spina bifida cystica. J Clin Pathol 18: Van Allen MI, Kalousek DK, Chernoff GF, Juriloff D, Harris M, McGillivray BC, Yong SL, Langlois S, MacLeod PM, Chitayat D, Friedman JM, Wilson RD, McFadden D, Pantzar J, Richtie S, Hall JD (1993) Evidence of multi-site closure of the neural tube in humans. Am J Med Genet 47: Vogter DM, Culberson JL, Schochet SS, Gabriele OF, Kaufman HH (1987) High spinal dysrhaphism. Case report of a complex cervical meningocele. Acta Neurochir (Wien) 84: Wu JK, Scott RM (1990) Myelopathy presenting decades after surgery for congenital cervical cutaneous lesions. Neurosurgery 27: