A Case of Junctional Neural Tube Defect Associated with a Lipoma of the Filum Terminale: a New Subtype of Junctional Neural Tube Defect?

CASE REPORT J Neurosurg Pediatr 21:601–605, 2018

A case of junctional neural tube defect associated with a lipoma of the filum terminale: a new subtype of junctional neural tube defect?

Simona Mihaela Florea, MD,1 Alice Faure, MD,2 Hervé Brunel, MD,3 Nadine Girard, MD, PhD,3 and Didier Scavarda, MD1

Departments of 1Pediatric Neurosurgery, 2Pediatric Surgery, and 3Neuroradiology, Hôpital Timone Enfants, Marseille, France

The embryological development of the central nervous system takes place during the neurulation process, which includes primary and secondary neurulation. A new form of dysraphism, named junctional neural tube defect (JNTD), was recently reported, with only 4 cases described in the literature. The authors report a fifth case of JNTD. This 5-year-old boy, who had been operated on during his 1st month of life for a uretero-rectal fistula, was referred for evaluation of possible spinal dysraphism. He had urinary incontinence, clubfeet, and a history of delayed walking ability. MRI showed a spinal cord divided in two, with an upper segment ending at the T-11 level and a lower segment at the L5–S1 level, with a thickened filum terminale. The JNTDs represent a recently classified dysraphism caused by an error during junctional neurulation. The authors suggest that their patient should be included in this category as the fifth case reported in the literature and note that this would be the first reported case of JNTD in association with a lipomatous filum terminale. https://thejns.org/doi/abs/10.3171/2018.1.PEDS17492 KEYWORDS junctional neurulation; junctional neural tube defect; spina bifida; dysraphism; spine; congenital

he central nervous system and vertebrae are formed or lipomas of the filum terminale.16 When there are altera- during the neurulation process that occurs early in tions present in both the primary and secondary neurula- the embryonic life and is responsible for the trans- tion we can find mixed dysraphisms that present with ele- Tformation of the flat neural plate into the neural tube (NT). ments from the 2 forms of neural tube defects (NTDs). Primary neurulation allows the formation of the brain and Recent papers have brought into discussion a new class the spinal cord down to the junction between the S-1 and of spinal dysraphisms that result from disturbances that S-2 segments, whereas secondary neurulation is respon- appear in the short period between the end of primary sible for the formation of the spinal cord segments distal to neurulation, marked by closure of the caudal neuropore, the S1–2 junction. and the beginning of secondary neurulation, consisting of Alteration of primary neurulation leads to malforma- the formation of the caudal cell mass.9,10,20,21 The phase be- tions of the spine and spinal cord known as open neural tween the end of primary neurulation and the beginning tube defects (ONTDs) in their most severe form or limited of secondary neurulation was termed junctional neurula- dorsal myeloschisis (LDM) in the lesser.6,18,19 tion,4 and it is thought to be responsible for maintaining a When the secondary neurulation process is impaired, continuity between the fully formed primary NT and the the result is either a defect of formation that leads to an secondary NT that begins to form. The alteration of this absent conus and a short spinal cord, a condition known transitional process leads to the apparition of a new dysra- as caudal agenesis;15 or a defect during its regression as we phism called junctional neural tube defect (JNTD).4,5,20,22 can see in terminal myelocystocele, ventriculus terminalis, In this paper, we report another case that we believe

ABBREVIATIONS JNTD = junctional neural tube defect; LDM = limited dorsal myeloschisis; MET = mesenchymal-epithelial transition; NSB = node-streak border; NT = neural tube; NTD = NT defect; ONTD = open NTD. SUBMITTED August 30, 2017. ACCEPTED January 3, 2018. INCLUDE WHEN CITING Published online March 30, 2018; DOI: 10.3171/2018.1.PEDS17492.

©AANS 2018, except where prohibited by US copyright law J Neurosurg Pediatr Volume 21 • June 2018 601

Unauthenticated | Downloaded 09/25/21 11:19 PM UTC S. M. Florea et al.

FIG. 3. Axial T1-weighted MR image obtained at the S-3 level, showing the lipomatous filum terminale black( arrow).

context of overall urinary incontinence. He had bilateral clubfeet for which he had undergone multiple interven- tions as well as orthopedic treatment such as immobiliza- tion in plaster splints. Examination of his lower extremities showed weakness of multiple muscles and atrophy. He had FIG. 1. Sagittal T2-weighted MR images of the thoracic-lumbar spine acquired the ability to walk with some delay, at about 2 showing the 2 segments of the spinal cord and the tissue band that con- years old, most probably related to the clubfeet. The patient nects them. The upper segment ends at the T11–12 level (white arrow), exhibited no cutaneous signs of occult spina bifida. whereas the lower segment is located at the L5–S1 level (gray arrow). Imaging can be included in this new class of dysraphisms described An MRI examination of the spinal cord was performed as JNTDs and discuss implications for the embryological (Figs. 1 and 2). This study showed a spinal cord divided in processes that lead to NTDs. two, with the upper segment ending abruptly at the T11–12 level and the lower segment located at the L5–S1 level and ending with a filum that was slightly thicker than normal, Case Report suggesting a slightly lipomatous nature (Fig. 3). The 2 seg- History and Presentation ments were connected by a 9.5-cm-long tract. The maxi- This 5-year-old boy was referred to our service for eval- mum thickness of the long tract connecting the 2 parts of uation of possible dysraphism. He had multiple malforma- the dysraphic spinal was 3 mm. A partial sacral agenesis tions. He had undergone surgical repair of a ureteral-rectal was observed, as well as other vertebral anomalies in the fistula at 1 month of age, and he had an anteriorly- dis cervical and thoracic regions. placed anus and partial sacral agenesis. He suffered from a hypertonic anal sphincter and was treated for constipa- Operation and Postoperative Course tion on a chronic basis. Urodynamic testing showed that Given the continence disturbances and the MRI ap- his bladder had a tendency for urinary retention, within the pearance of the filum, we suspected that the patient had a

FIG. 2. Axial T2-weighted MR images. A: Image obtained at the T-11 level, showing the upper segment of the spinal cord (arrow). B: Image obtained at the L-2 level, showing the band (arrow) that connects the 2 segments of the spinal cord. C: Image obtained at the L-5 level, showing the lower segment of the spinal cord (arrow).

602 J Neurosurg Pediatr Volume 21 • June 2018

Unauthenticated | Downloaded 09/25/21 11:19 PM UTC S. M. Florea et al.

severe forms of ONTDs such as anencephaly, encephalo- celes, or myelomeningoceles to less severe malformations, known as LDMs.6,18,19 Secondary neurulation results in the formation of the secondary NT and it is essentially different from the primary neurulation. During this phase, the caudal cell mass passes through a process called mesenchymal-epithelial transition (MET)2,3,7,8,11,13,24,27,28 and is eventually responsible for the formation of the spinal cord segment from S-2 distally, together with the filum terminale, but also for the development of most of the hind gut, lower genitourinary tract, and sacrococcygeal osseous elements.9,10,12,13 Impair- ment of this process results in caudal agenesis or regression syndromes, which are frequently associated with visceral and vertebral malformation due to the common origin of these entities.5,10,11,14–17,21,31 Recently a third category of NTDs, named junctional neural tube defects (JNTDs), has been proposed. These defects that are thought to originate from alterations of the transitional process between primary neurulation, marked by the closing of the posterior neuropore and the beginning of secondary neurulation.7 In their 2014 article, Dady et al.4 described the mechanisms by which the continuity FIG. 4. Sagittal T2-weighted images obtained 3 months after surgery. of the 2 segments of the spinal cord resulting from the well-known neurulation processes is maintained. Accord- ing to their work, the cells located between the Hensen’s tethered cord and proposed surgical sectioning of the lipo- node and the cranial end of the primitive streak, which matous filum. Intraoperatively we found a thickened filum they named “node-streak border” (NSB) are influenced in terminale, with a macroscopically lipomatous aspect, but their activity by the expression of the SOX-2 protein. Dady we observed that there was no tension in the filum during and colleagues demonstrated that the SOX-2–positive cells its manipulation. Despite the lack of tension, we decided to located dorsally will move dorsolaterally and be part of perform the sectioning as planned and analyze the resect- ed tissue microscopically, given that the appearance was the primary neural tube, while the cells that become SOX- compatible with a filum terminale lipoma. At 3 months af- 2 positive during the MET process will migrate caudally ter the surgery, there was no improvement concerning the and participate at the formation of the secondary NT, dur- urinary troubles, and there was no change on MRI with ing secondary neurulation. The SOX-2 positivity of the regard to the position of the lower segment of the spinal NSB cells maintains a connection between the dorsolateral cord (Fig. 4). These elements rather suggest that there was and the ventral cell populations. The authors also showed no tethered cord from the beginning, and that there is no the importance that the basement membrane of the NSB functional connection between the 2 spinal cord segments. cells has in allowing an accurate fibronectin distribution along it, facilitating coupling between the 2 cell populations. Insults during these molecular processes, along with Discussion the normal development of the embryo, might explain the Until recently, when referring to NTDs we would cat- NTD described as junctional neural tube defect (JNTD). egorize the malformation as either a primary or secondary In 2017, Eibach et al.5 reported 3 cases of patients pre- neurulation defect. Primary neurulation consists of mor- senting with 2 spinal cord segments that were completely phogenetic movements leading to the shaping and bending separated functionally and were connected anatomically of the neural plate and results in forming the brain and the by a narrow band that did not contain neural tissue. Clini- spinal cord to the junction between the S-1 and S-2 seg- cally, all 3 patients had fecal and urinary incontinence, 2 of ments. The process of shaping the neural plate includes the patients presented with clubfeet, and all presented with the following 3 types of morphogenetic movements, which spinal deformity. In January 2017, Schmidt et al.22 reported convert the oval neural plate into the narrow spinal plate another case of JNTD. Their case involved a newborn who and the wider brain plate: apicobasal thickening, transverse appeared to have normal movements in the lower extremi- narrowing, and longitudinal lengthening.3,13,23,25,26,29 Dur- ties and had no cutaneous stigmata for dysraphism, but ing the bending of the neural plate, a process that occurs whose MRI findings were similar to those described in the after the shaping of the neural plate, the neural primor- cases reported by Eibach et al. To our knowledge, these are dium transforms into the V-shaped neural groove through the only 4 cases of JNTDs reported in the literature so far. medial furrowing,1,13,30 and then the folding of the groove We believe that our case may be included in the same turns it into the neural tube, by allowing the neural plates category of JNTDs as the other 4 cases, as our patient to converge and fuse on the posterior midline.1,13 When presented with similar clinical and imaging findings, as primary neurulation is altered, the fetuses will have clo- summarized in Table 1. The distinguishing feature of our sure defects of the dorsal midline, ranging from the most case is the association of a lipomatous filum terminale,

J Neurosurg Pediatr Volume 21 • June 2018 603

Unauthenticated | Downloaded 09/25/21 11:19 PM UTC S. M. Florea et al.

TABLE 1. Characteristics of reported cases of JNTB Authors Age (yrs), Associated & Year Sex Clinical Examination Imaging Malformations Procedure Eibach 8, M Delayed walking; L-5, S-1 muscle Upper spinal cord segment T-12 & L-3 hemi- Partial resection of filum terminale et al., weakness & muscle atrophy; pri- ending at L-1, lower segment vertebrae; (HP: normal structure); intraoper- 20175 mary urinary & fecal incontinence; at L3–4 level, connected by partial sacral ative electrophysiological studies hypesthesia L-5, S-1; anesthesia a 2-mm-thick band; normal agenesis; showing that the 2 segments of S2–4; no cutaneous signs appearance of filum scoliosis the spinal cord correspond to the primary & secondary NT, w/ no functional connection btwn them 13, F Delayed walking; L-5, S-1 muscle Upper spinal cord segment end- Lumbosacral No surgical procedure performed weakness, muscle atrophy; pri- ing at T12–L1, lower segment vertebral mary urinary & fecal incontinence; at S1–2 level, connected by segmentation hypesthesia L-5, S-1; anesthesia a band; normal appearance failure; scolio- S2–4; no cutaneous signs of filum sis; clubfeet 30, F Delayed walking; L-5, S-1 muscle Upper spinal cord segment Partial sacral No surgical procedure performed weakness, muscle atrophy; pri- ending at T-11, lower segment agenesis, mary urinary & fecal incontinence; at L5–S1 level, connected by scoliosis, hypesthesia L-5, S-1; anesthesia a 3-mm thick band; normal clubfeet S2–4; no cutaneous signs appearance of filum Schmidt Newborn, No cutaneous signs; normal move- Upper spinal cord segment end- Anorectal No surgical procedure performed et al., M ments in upper & lower extremi- ing at T-11, lower segment at atresia, 2017 ties L-2 level, connected by a band; partial sacral normal appearance of filum agenesis Present 5, M Delayed walking; L-5, S-1 muscle Upper spinal cord segment end- Ureteral-rectal Partial resection of filum terminale case weakness, muscle atrophy; ing at T11–12, lower segment fistula; partial (HP: lipomatous filum) primary urinary incontinence, at L5–S1 level, connected by sacral agen- hypertonic anal sphincter; no a 3-mm-thick band; thick filum esis; clubfeet cutaneous signs terminale HP = histopathology. as confirmed histopathologically, without a tethered cord. References This aspect might suggest either that the mechanisms im- plicated in junctional neurulation also have a certain role 1. Alvarez IS, Schoenwolf GC: Expansion of surface epithelium provides the major extrinsic force for bending of the neural in secondary neurulation or that the factors that led to plate. J Exp Zool 261:340–348, 1992 the malformation may interfere in secondary neurulation 2. Catala M, Teillet MA, Le Douarin NM: Organization and if its duration is sufficient. Also, simply considering that development of the tail bud analyzed with the quail-chick the embryological processes of junctional neurulation and chimaera system. Mech Dev 51:51–65, 1995 secondary neurulation are not initiated consecutively, but 3. Colas JF, Schoenwolf GC: Towards a cellular and molecular rather concomitantly, might also be an explanation for the understanding of neurulation. Dev Dyn 221:117–145, 2001 association between the JNTD and the lipomatous filum 4. Dady A, Havis E, Escriou V, Catala M, Duband JL: Junction- terminale in our patient. Other elements that suggest this al neurulation: a unique developmental program shaping a possibility are the associated malformations that are ex- discrete region of the spinal cord highly susceptible to neural plained by alteration of the secondary neurulation process tube defects. J Neurosci 34:13208–13221, 2014 5. Eibach S, Moes G, Hou YJ, Zovickian J, Pang D: Unjoined in the previously reported cases, as we can see in the cau- primary and secondary neural tubes: junctional neural tube dal regression NTD. defect, a new form of spinal dysraphism caused by distur- bance of junctional neurulation. Childs Nerv Syst 33:1633– Conclusions 1647, 2017 JNTDs were recently classified as a third category of 6. Eibach S, Moes G, Zovickian J, Pang D: Limited dorsal my- dysraphism that is thought to appear as a result of an error eloschisis associated with dermoid elements. Childs Nerv during the junctional neurulation, a transitional process Syst 33:55–67, 2017 that takes place between the primary and the secondary 7. Hughes AF, Freeman RB: Comparative remarks on the development of the tail cord among higher vertebrates. J Em- neurulation. We reported a fifth case of JNTD, as dem- bryol Exp Morphol 32:355–363, 1974 onstrated by the clinical and MRI findings, that is similar 8. Lowery LA, Sive H: Strategies of vertebrate neurulation and to the other cases previously reported. Although in our a re-evaluation of teleost neural tube formation. Mech Dev patient we also found a lipomatous filum, we believe that 121:1189–1197, 2004 there was no tethered cord associated with it and that the 2 9. Müller F, O’Rahilly R: The development of the human brain, spinal cord segments were functionally separate. the closure of the caudal neuropore, and the beginning of

604 J Neurosurg Pediatr Volume 21 • June 2018

Unauthenticated | Downloaded 09/25/21 11:19 PM UTC S. M. Florea et al.

secondary neurulation at stage 12. Anat Embryol (Berl) rearrangement and division in shaping of the avian neural 176:413–430, 1987 plate. Development 106:427–439, 1989 10. Müller F, O’Rahilly R: The primitive streak, the caudal emi- 27. Schoenwolf GC, Delongo J: Ultrastructure of secondary neu- nence and related structures in staged human embryos. Cells rulation in the chick embryo. Am J Anat 158:43–63, 1980 Tissues Organs 177:2–20, 2004 28. Schoenwolf GC, Smith JL: Mechanisms of neurulation. 11. Nievelstein RA, Hartwig NG, Vermeij-Keers C, Valk J: Em- Methods Mol Biol 136:125–134, 2000 bryonic development of the mammalian caudal neural tube. 29. Schoenwolf GC, Smith JL: Mechanisms of neurulation: Teratology 48:21–31, 1993 traditional viewpoint and recent advances. Development 12. O’Rahilly R, Müller F: Neurulation in the normal human 109:243–270, 1990 embryo. Ciba Found Symp 181:70–89, 1994 30. Smith JL, Schoenwolf GC: Cell cycle and neuroepithelial 13. Ӧzek MM, Cinalli G, Maixner VJ (eds): Spina Bifida. Man- cell shape during bending of the chick neural plate. Anat agement and Outcome. Milan: Springer, 2008, pp 19–39 Rec 218:196–206, 1987 14. Padmanabhan R: Etiology, pathogenesis and prevention of 31. Yang HJ, Lee DH, Lee YJ, Chi JG, Lee JY, Phi JH, et al: neural tube defects. Congenit Anom (Kyoto) 46:55– 67, Secondary neurulation of human embryos: morphological 2006 changes and the expression of neuronal antigens. Childs 15. Pang D: Sacral agenesis and caudal spinal cord malforma- Nerv Syst 30:73–82, 2014 tions. Neurosurgery 32:755–779, 1993 16. Pang D, Zovickian J, Lee JY, Moes GS, Wang KC: Terminal myelocystocele: surgical observations and theory of embryo- genesis. Neurosurgery 70:1383–1405, 2012 17. Pang D, Zovickian J, Moes GS: Retained medullary cord in humans: late arrest of secondary neurulation. Neurosurgery Disclosures 68:1500–1519, 2011 The authors report no conflict of interest concerning the materi- 18. Pang D, Zovickian J, Oviedo A, Moes GS: Limited dorsal als or methods used in this study or the findings specified in this myeloschisis: a distinctive clinicopathological entity. Neuro- paper. surgery 67:1555–1580, 2010 19. Pang D, Zovickian J, Wong ST, Hou YJ, Moes GS: Limited Author Contributions dorsal myeloschisis: a not-so-rare form of primary neurulation defect. Childs Nerv Syst 29:1459–1484, 2013 Conception and design: Scavarda, Florea. Acquisition of data: 20. Saitsu H, Yamada S, Uwabe C, Ishibashi M, Shiota K: Devel- Scavarda, Florea, Brunel, Girard. Analysis and interpretation of opment of the posterior neural tube in human embryos. Anat data: Scavarda, Florea, Faure, Brunel. Drafting the article: Florea. Embryol (Berl) 209:107–117, 2004 Critically revising the article: Scavarda, Faure, Girard. Approved 21. Saraga-Babic M, Krolo M, Sapunar D, Terzic J, Biocic M: the final version of the manuscript on behalf of all authors: Scav- Differences in origin and fate between the cranial and caudal arda. Administrative/technical/material support: Scavarda. Study spinal cord during normal and disturbed human develop- supervision: Scavarda. ment. Acta Neuropathol 91:194–199, 1996 22. Schmidt C, Voin V, Iwanaga J, Alonso F, Oskouian RJ, To- Supplemental Information pale N, et al: Junctional neural tube defect in a newborn: report of a fourth case. Childs Nerv Syst 33:873–875, 2017 Previous Presentations 23. Schoenwolf GC: Cell movements driving neurulation in The abstract for this article was presented as an oral presenta- avian embryos. Dev Suppl 2 (Suppl 2):157–168, 1991 tion at the annual meeting of the Société de Neurochirurgie de 24. Schoenwolf GC: Histological and ultrastructural studies Langue Française, RAP 2017, in Paris, November 27–29, 2017. of secondary neurulation in mouse embryos. Am J Anat 169:361–376, 1984 Correspondence 25. Schoenwolf GC, Alvarez IS: Role of cell rearrangement in axial morphogenesis. Curr Top Dev Biol 27:129–173, 1992 Didier Scavarda: Hôpital Timone Enfants, Marseille, France. 26. Schoenwolf GC, Alvarez IS: Roles of neuroepithelial cell [email protected].

J Neurosurg Pediatr Volume 21 • June 2018 605

Unauthenticated | Downloaded 09/25/21 11:19 PM UTC