Late rupture of a myelomeningocele sac devoid of neural placode associated with good functional outcome 2 years after postnatal repair: illustrative case

Abstract

BACKGROUND

Neurological deficits in myelomeningocele are driven by a three-hit paradigm: primary myelodysplasia, intrauterine chemotoxic amniotic fluid exposure, and mechanical traction on the neural placode. How sac integrity and rupture timing influence these progressive pathophysiological hits remains poorly understood.

OBSERVATIONS

A neonate presented with a prominent S1 myelomeningocele sac that remained intact until approximately 35 weeks’ gestation. Rather than a distinct neural placode, postnatal repair at 36+3 weeks revealed a hairpin loop of functional nerve roots within the myelomeningocele sac that entered and exited via the dorsal defect. A mild Chiari type II malformation was identified only on postnatal imaging. Following postnatal repair, the child demonstrated not only excellent lower extremity motor function as expected, but also excellent urinary sphincter function with no hindbrain herniation or ventriculomegaly 2 years postrepair.

LESSONS

This case illustrates a possible “best of both worlds” clinical scenario in which the persistent hydrostatic pressure of the intact sac protected against early hindbrain herniation and direct toxicity, while the structural absence of a neural placode within the sac reduced mechanical stretch injury. Finally, rupture of the myelomeningocele sac only in late gestation minimized chemical toxicity to neural elements. Overall, this case highlights the multifaceted and varied nature of myelomeningocele pathophysiology.

https://thejns.org/doi/abs/10.3171/CASE26739

Myelomeningocele is the most severe form of spina bifida and results from the failure of neural tube closure, leading to the protrusion of the meninges and spinal cord through a defect in the vertebral arches. The resulting neurological deficits encompass a broad spectrum of motor, sensory, and bowel or bladder dysfunction dictated primarily by the anatomical level of the lesion.^1–3 In addition to spinal deficits, nearly 80% of affected neonates present with hydrocephalus, requiring surgical CSF diversion in approximately 75% of cases.^4,5 Chiari type II malformations in myelomeningocele are associated with poor outcomes and mortality.⁶ These central nervous system anomalies are frequently tied to significant, lifelong cognitive impacts, including deficits in executive function and language that ultimately reduce overall quality of life.^7,8

While the severity of these manifestations has historically been attributed to the baseline spinal defect, mounting evidence suggests that the final neurological deficit cannot be explained by primary myelodysplasia alone.^9–11 The widely accepted two-hit hypothesis posits that the initial neurodevelopmental failure is compounded by a secondary hit: the prolonged exposure of neural elements to the chemotoxic intrauterine environment.^10,11 More recently, an extension to a three-hit hypothesis has been proposed, identifying a tertiary source of trauma from mechanical traction and hydrostatic forces acting directly on the neural placode.¹²

Here, we present a unique case of lumbosacral myelomeningocele characterized by two unusual features: 1) a fluid-filled sac at L5–S1 containing functional neural elements that remained unruptured until late in gestation, and 2) the absence of a distinct neural placode. Two years following postnatal repair, the child demonstrates excellent lowe extremity motor and urinary function without ventriculomegaly or persistent hindbrain herniation.

This case provides clinical insight into the three-hit framework, illustrating how late sac rupture and the atypical absence of a neural placode can shield the central nervous system from both chemotoxic exposure and mechanical traction, thereby contributing to a favorable neurological outcome.

Illustrative Case

A 37-year-old G11P4A6 woman was referred to our neurosurgery clinic at 24 weeks’ gestation after routine prenatal ultrasonography identified a fetal neural tube defect. Her first child was born preterm at 27 weeks, while her subsequent children were born at term. Her past medical history was significant for thyroid cancer for which she underwent hemithyroidectomy, hypertension, a BMI of 39, cholecystectomy, anxiety, posttraumatic stress disorder, and a cardiac murmur from childhood that had resolved. Her medications included bupropion, levothyroxine, liothyronine, fluoxetine, insulin, and prenatal vitamins. She was taking folic acid both prenatally and early in pregnancy. Her personal and family history was negative for neural tube defect. Her current pregnancy was marked by early gestational diabetes that was controlled with diet but was otherwise uneventful.

Prenatal Course

Ultrasonography performed at 19+4 weeks was suspicious for a meningocele (Fig. 1A). Gestational MRI of the fetus at 21+5 weeks showed a lumbosacral fluid-filled sac with neural tissue extending into the defect toward the dorsal surface of the sac (Fig. 1B). No hindbrain herniation or ventriculomegaly was noted. Aside from a right-sided grade III hydronephrosis, the remainder of the fetal survey was normal. Amniocentesis revealed a normal karyotype. Repeat MRI performed at 28+0 weeks showed interval enlargement of the fluid-filled sac (Fig. 1C), but still no associated Chiari malformation or ventriculomegaly.

FIG. 1.

Prenatal (A–D) and postnatal (E–I) images showing pertinent findings. A: Fetal routine ultrasound performed at 19+4 weeks of gestational age showing a meningocele (arrow). B: Fetal MR image demonstrating a fluid-filled sac measuring 28 × 16 × 32 mm at the lumbosacral area, with neural tissue extending into the defect (arrow). C: Follow-up fetal MR image demonstrating progression in size of the fluid-filled sac (arrow). D: Fetal ultrasound showing rupture of the sac (arrow). E: Sagittal T2-weighted MR image of the spine showing the low-lying conus at approximately L3–4, as well as a dorsal defect at the level of S1 with herniation of neural tissue into the deflated sac. F: Axial cuts (corresponding levels shown by the red dashed lines in panel E) demonstrating the relation between the neural elements and the deflated sac, with F3 showing herniation of neural tissue into the sac at S1 (arrow). G and H: Axial cuts of MR images of the head showing a small left grade I germinal matrix hemorrhage (arrow). I: T1-weighted midsagittal MR cut of the head showing a mild Chiari type II malformation (arrow).

Following discussion in the multidisciplinary Fetal Diagnosis and Treatment group, the patient was not recommended for intrauterine repair given her elevated BMI, previous history of preterm birth, and low level of the defect, all of which were consistent with the Management of Myelomeningocele Study (MOMS) exclusion criteria.¹³ The atypical pattern of neural tissue within the sac also contributed to this decision. An elective caesarean section at 36+3 weeks in a high-risk unit with planned neonatal involvement was recommended.

At 35 weeks, a repeat ultrasound revealed rupture of the sac with minimal residual fluid inside (Fig. 1D). Considering the lack of hindbrain herniation and ventriculomegaly seen on imaging, we opted to wait until the planned delivery.

Immediate Postpartum Course

At 36+3 weeks, a male neonate was delivered via caesarean section. A deflated approximately 6.7 × 8–cm skin-covered sac was noted at the lumbosacral region with a cleft in the right paramedian aspect (Fig. 2A). CSF was seen leaking through the defect in the sac, and we could visualize nervous tissue within. The baby was kept prone, and the lesion was covered with sterile dressing. We kept the baby on prophylactic cefotaxime and ampicillin until surgical repair approximately 48 hours later.

FIG. 2.

Intraoperative images of myelomeningocele repair. A: Deflated approximately 6.7 × 8–cm skin-covered sac with a cleft in the right paramedian aspect. B: Photograph showing nerve roots (elevated by instrument and marked by blue dotted arrow) exiting the dorsal defect, forming a hairpin loop and attaching to the inner wall of the sac, and then reentering the defect. C: Detachment of neural elements (blue dotted arrow) from the sac and covered with hemostatic material. D: Exposure of L5 lamina (arrow). E:Laminectomy and exposure of the normal thecal sac above the defect (arrow). F: Dural opening and exposure of the cauda equina and conus above the defect. Dotted white lines indicate the tacked-up dural edges exposing the conus (arrow). G: Dissection of arachnoid around the nerve roots and identification and cutting of the filum terminale under a microscope. H: Dissection of the junctional zone (white dotted line) and creation of a flap (blue arrows) to re-create the thecal sac containing the herniated nerve roots. I: Trimming of excess skin and closure.

On examination, the head circumference was at the 15th percentile and there was a soft anterior fontanelle. The baby was active and reactive with spontaneous eye opening. There was a positive suck and grasp reflex, and he moved both upper and lower extremities equally and spontaneously. No orthopedic abnormalities were noted. Bilateral plantar and toe flexion as well as perineal reflex were present. The baby was passing urine and stool spontaneously.

MRI of the head and spine was performed prior to surgical management (Fig. 1E–I). Imaging showed a dorsal defect at L5–S1 with herniation of neural tissue into the deflated sac (Fig. 1E and F). There was evidence of tethering and a low-lying conus at approximately L3–4. There was progression in the right-sided hydronephrosis with a new low-grade left-sided hydronephrosis also noted. MRI of the head showed a mild Chiari type II malformation and small grade I germinal matrix hemorrhage (Fig. 1G–I) that were new since the previous MRI examination.

Surgical Repair

Surgical repair was performed at approximately 48 hours of life. Rather than the typical exposed neural placode with ventrally oriented nerve roots, we observed a single bundle of nerves exiting the dorsal spinous defect, attaching rostrally to the inside of the sac, forming a hairpin loop, and reentering the defect (Fig. 2B–I). Given this ­hairpin configuration and the late rupture of the myelomeningocele sac, we opted to use direct nerve root stimulation to offer intraoperative guidance. Stimulation of this bundle of nerves induced bilateral ­dorsiflexion, plantar flexion, and toe flexion, indicating functional tissue. Closure of the defect was uneventful, although additional intradural space was required to reintegrate the hairpin loop of nerves into the canal. This was accomplished by dissecting the junctional zone more laterally.

Postoperative Course

Postoperatively, the baby retained similar lower extremity motor function. The perineal reflex was preserved, and he was still able to spontaneously void. Head ultrasound showed stable ventricular size. Prophylactic antibiotic dosing was continued for 5 days, and the child was discharged home on postoperative day 8 in good condition. An ultrasound of the bladder revealed grade III hydronephrosis of the right kidney, and the child was started on prophylactic antibiotics.

Follow-Up

On follow-up in the spina bifida clinic 1 year later, head ultrasound showed no ventriculomegaly. The baby showed grossly normal sensation and motor development for his age and was crawling, pulling to stand, and cruising using furniture with good movement of both feet. He was able to self-void without a need for catheterization. He had slightly worse grade III hydronephrosis of the right kidney. A mercaptoacetyltriglycine renal scan revealed normal kidney drainage and function. A cystourethrogram demonstrated left grade III vesicoureteral reflux. He did not have any urinary infections; thus, prophylactic antibiotics were stopped. The child was deemed stable from a urological perspective.

At 2 years of age, a urodynamic study demonstrated normal bladder capacity and compliance for age, with baseline pressures of 25–30 cm H₂O and a leak point pressure of 64 cm H₂O. The child was emptying his bladder spontaneously and had no history of urinary infections. The child was also ambulating independently and climbing on furniture. While the child had good quadriceps and hamstring strength bilaterally, inversion and weakness of the right ankle was noted. Repeat imaging showed a new sacral arachnoid cyst, lumbar syringomyelia, and tethered cord (Fig. 3A and B). There was no hindbrain herniation. The early lumbar hydromyelia was attributed to secondary tethering given the absence of hindbrain herniation. As the original repositioning of the hairpin nerve roots in the canal may have altered CSF flow and caused mass effect on the distal cauda equina, we opted for release of the tethered cord and placement of a cystosubarachnoid shunt. Postoperative imaging demonstrated restoration of the posterior CSF space, collapse of the arachnoid cyst, and modest decrease in size of the hydromyelic cavity (Fig. 3C–E). Following surgery, the child showed improved right ankle strength and gait and is progressing well in toilet training.

FIG. 3.

Preoperative (A and B) and postoperative (C–E) images. A:Sagittal T2-weighted MR image of the spine showing a newly developed 2.3 × 1.6 × 1.9–cm sacral cystic lesion at S2–4. The neural elements were deviated to the right and tethered. A syrinx was also seen from T12 to L5. B: Axial MR slices, with corresponding levels shown by the red dashed lines in panel A. The patient underwent release of the tethered cord and placement of a cystosubarachnoid shunt. C and D: T2-weighted sagittal MR slices showing reduction of the cystic lesion, mild improvement of the syrinx, and placement of the cystosubarachnoid shunt catheter (arrow). E:Axial slices, with corresponding levels indicated by the red dashed lines in panel D.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

In the present case, we demonstrate excellent functional outcomes following postnatal repair of a myelomeningocele characterized by an atypical absence of the neural placode within the sac and delayed exposure of neural tissue to amniotic fluid. After review of the literature, and to our knowledge, there is no similar previously reported case.

While the low anatomical level of the defect (S1) was likely the principal determinant of the favorable outcome, we hypothesize that the patient’s favorable neurological outcome was also due in large part to the strictly limited duration of exposure of neural tissue to amniotic fluid, the prevention of hindbrain herniation by maintenance of CSF hydrostatic forces, and the absence of a distinct neural placode susceptible to mechanical traction.

Animal and human fetal studies suggest that chronic exposure of neural tissue to the intrauterine environment induces chemical neuronal degradation and astrogliosis that directly correlate with impaired postnatal motor function.^9,14,15 Similarly, in vitro models have demonstrated accelerated neuronal cell death in spinal cord tissue exposed to amniotic fluid.¹⁶ In the present case, the intact myelomeningocele sac likely shielded the spinal cord from this chemotoxic environment until 35 weeks’ gestation, at which point the more mature nerve roots may have been more resilient to chemical insult. Furthermore, because the neural placode was uniquely absent from the sac itself, only minimal neural tissue was subjected to direct amniotic fluid exposure even following sac rupture.

The myelomeningocele sac remained intact until approximately 35 weeks, with hindbrain herniation appearing only on postnatal imaging. An intact sac maintains CSF hydrostatic pressure within the spinal canal, potentially reducing downward displacement of the hindbrain—a critical prognostic factor for functional impairment and mortality.^6,17 This concept is supported by a retrospective MRI study comparing myelomeningocele to myelocele (where the spinal cord remains in the canal and the placode is exposed flush with the skin), which found a significantly higher incidence of severe Chiari type II malformations in myelocele cases due to the unmitigated egress of CSF.¹⁸ Similarly, Nagaraj et al. also identified an inverse relationship between sac size and the degree of hindbrain herniation, reinforcing the concept that a persistent sac protects against Chiari type II progression.¹⁸

Conversely, a post hoc analysis of the MOMS trial data suggested that the absence of a sac positively correlated with independent ambulation speeds following prenatal repair,¹⁹ while Oliver et al. noted an association between larger sac size and higher rates of talipes.¹² These findings led the authors to propose that a prominent, fluid-filled sac introduces another mechanical hit via stretch injury of the contained neural placode. This mechanical theory is consistent with the lower rates of talipes observed in myelocele compared to myelomeningocele, as myeloceles lack a distending sac exerting traction on nerve roots.

Our case represents a unique clinical scenario in which both pathological mechanisms may have been attenuated. The persistence of a fluid-filled sac for the majority of gestation likely maintained CSF hydrostatic pressure and averted early hindbrain herniation, while the structural absence of a neural placode within that sac precluded mechanical stretch injury. Together with the limited duration of amniotic fluid exposure, these anatomical factors likely contributed synergistically to preserve neurological and urinary sphincter function.

Lessons

Overall, this case illustrates the complex interplay between amniotic fluid toxicity, CSF hydrostatic forces, and mechanical traction in influencing outcome in myelomeningocele.

Timing of rupture of the myelomeningocele sac is a key determinant of neurological outcome by influencing the duration of exposure of neural elements to the chemotoxic intrauterine milieu, highlighting the importance of early in utero repair of myelomeningocele in eligible patients. At the same time, the presence of an intact myelomeningocele sac may also be protective against a Chiari type II malformation by offering hydrostatic force against hindbrain herniation. Conversely, a distended myelomeningocele sac may also induce a stretch injury on the neural placode, leading to worse motor outcomes. These seemingly opposing mechanisms emphasize the importance of considering each patient’s unique anatomical characteristics when evaluating prognosis and treatment.

While it is difficult to disentangle the effects of a defect at a low anatomical level (S1) from the chemical and mechanical protection afforded by an intact sac, as well as the lack of a distinct neural placode, the exceptional preservation of urinary and lower extremity motor function at 2 years of age, together with the absence of a prenatally apparent Chiari type II malformation, suggests that these unique morphological features acted synergistically to mitigate progressive, secondary neurological injury in utero. Further studies examining the effect of sac morphology and rupture timing on long-term postrepair functional outcomes are warranted.

Despite successful surgical repair and a favorable initial outcome, close multidisciplinary surveillance remains essential, as urinary, bowel, and sexual dysfunction may manifest later in life.^20–22 Although overall mortality among individuals with spina bifida has improved over recent decades,²³ urological complications remain a primary driver of premature death in adult patients.²⁴ Careful lifelong surveillance of bladder function and maintenance of adequate bladder emptying are therefore critical to minimizing the risk of hydronephrosis, renal deterioration, and subsequent renal failure.

Acknowledgments

We thank the family of the patient who agreed to take part in the case report. Gemini 3.6 was used for text editing.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: JP Farmer, Lam, Cattani. Acquisition of data: Lam. Analysis and interpretation of data: JP Farmer, Lam, DL Farmer. Drafting the article: JP Farmer, Lam, Guberman, Sahlas. Critically revising the article: JP Farmer, Lam, Guberman, Sahlas, Veilleux, Cattani. Reviewed submitted version of manuscript: JP Farmer, Lam, Guberman, Sahlas, Veilleux, DL Farmer. Approved the final version of the manuscript on behalf of all authors: JP Farmer. Administrative/technical/material support: JP Farmer, Lam, Altuhayni, DL Farmer. Study supervision: Altuhayni, Cattani.

Correspondence

Jean-Pierre Farmer: Montréal Children’s Hospital, Montréal, QC, Canada. jean-pierre.farmer@mcgill.ca.