Published online October 1, 2004
PEDIATRICS
Vol. 114
No. 4
October 2004, pp.
e527-e531
(doi:10.1542/peds.2003-0496-F)
Presacral Neuroenteric Fistula in a Newborn Presenting With an Epidural Abscess: Case Report and Review of the Literature
Balsam Darwish, FRACS*,
Thorsten V. Stanley, FRCP
,
Agadha Wickremesekera, FRACS*,
Brendon Bowkett, FRACS
and
Mark Leadbitter, FRACR||
* Departments of Neurosurgery
Paediatrics
Paediatric Surgery
|| Radiology, Wellington Hospital and Wellington School of Medicine and Health Sciences, Wellington, New Zealand
 |
ABSTRACT
|
|---|
We describe a newborn infant (<24 hours of age) who presented
with mild swelling on the back and buttocks attributable to
a neuroenteric fistula complicated by an epidural parasacral
abscess infected with mixed coliforms. Epidural abscesses in
infancy are extremely rare, and one has not been observed previously
in the newborn period. The infant was surprisingly mildly affected.
Prompt intervention led to an excellent outcome. Coliforms may
colonize the infant gut in <24 hours, even in the developed
world. Unexplained swellings on the backs of infants should
lead to a search for underlying malformations and an early surgical
review, which is best conducted with a combined pediatric surgical
and neurosurgical approach.
Key Words: newborn infant • neuroenteric fistula • epidural abscess • coliform • meningitis
Abbreviations: MRI, magnetic resonance imaging • SEA, spinal epidural abscess
Neuroenteric cysts (fistulas) are rare, with <30 cases being described in the pediatric literature. Epidural abscesses presenting in infancy are even rarer, with the youngest patient reported being 20 days of age.1 An epidural abscess complicating a congenital neuroenteric cyst appears not to have been reported previously.
|
CASE REPORT
|
|---|
A <1-day-old boy was admitted to the neonatal unit at a peripheral
hospital with irritability, lethargy, and poor feeding. He had
been delivered at 42 weeks of gestation through normal vaginal
delivery and weighed 3250 g. His mother had an older healthy
child and had remained in good health during this pregnancy.
She admitted to smoking cigarettes and occasionally drinking
alcoholic drinks. Results of ultrasound scanning performed in
the first trimester had been reported as normal and consistent
with dates. The infant’s condition at birth was excellent,
and he was allowed home, breastfeeding successfully, at 12 hours
of age. In the examination, the infant was mildly jaundiced
and lethargic. He had cool extremities, with a capillary return
of 4 seconds, a heart rate of 150 beats per minute, a respiratory
rate of 68 breaths per minute, and a skin temperature of 36.6°C.
There was no sign of respiratory difficulty. The patient’s
anterior fontanelle was not distended, and his posture was normal.
His abdomen was soft, without organomegaly. Movements were normal
in all limbs, with normal tone and reflexes. The patient’s
anus was in a normal position, and he passed stools without
difficulty. His back was normal, with no sign of sinuses or
fistulas. Blood and urine cultures were obtained. Blood tests
showed a C-reactive protein level of 25 µg/mL (normal:
<10 µg/mL) and a white blood cell count of 10 300 cells
per mm
3 (neutrophils: 8000 cells per mm
3). Results of a chest
radiograph were normal. Lumbar punctures were performed as part
of the septic screening and on 2 occasions revealed frank yellow-green
pus, which showed large numbers of pus cells and mixed flora
with Gram-staining and subsequently showed heavy growth of
Escherichia coli and enterococci. A provisional diagnosis of mixed-organism
meningitis was made, and the infant began broad-spectrum antibiotic
treatment and was transferred to the neonatal intensive care
unit of Wellington Hospital for additional investigations and
treatment.
At the time of arrival, the infant’s posture was noted to be abnormal; he was lying somewhat hyperextended and was irritable during handling. There was an indistinct but definite swelling over the lower lumbar spine, which appeared tender (Fig 1). There was some erythema over the right posterior lower trunk and buttocks, extending into the inguinal region, with deviation of the gluteal cleft to the left; the anus was thought to be mildly displaced anteriorly. There was a tender, mildly fluctuant swelling (5 cm x 2 cm) over the superomedial buttock area on the right (Fig 2), which appeared to be separate from the swelling over the lumbar spine. The infant exhibited mild tenderness in the right renal angle with deep palpation. A provisional diagnosis of spinal epidural abscess (SEA) was made, but the gluteal swelling was unexplained. The infant underwent ultrasound scanning of the lumbar spine and urgent magnetic resonance imaging (MRI) of the pelvis. On ultrasound scans, the conus was at L1. The nerve roots of the cauda equina were bunched together, with reduced pulsation, and a paucity of surrounding cerebrospinal fluid was noted. On MRI scans, a presacral cystic mass measuring 5 cm x 2 cm was confirmed; this mass was pushing the rectum and bladder forward (Figs 3 and 4). The mass extended through the greater sciatic notch on the right, into the right buttock region (Fig 3). Air was clearly visible within the mass. The lesion extended into the epidural space of the spinal canal, with air within the extradural space (Figs 5 and 6) reaching the upper cervical spine. Gadolinium enhancement of the tissue in the epidural space was noted, in keeping with local inflammatory changes. The bony elements of the sacrum were intact. The findings strongly suggested fistulous communication between the rectum, the presacral cyst, and the epidural space. Results of MRI scans of the brain were normal.
View larger version (130K):
[in this window]
[in a new window]
|
Fig 3. Axial, T2-weighted, fat-saturation, MRI scan of the pelvis, showing the rectum and bladder pushed forward (transparent arrows) and a mass extending into the right buttock (solid arrow).
|
|
View larger version (150K):
[in this window]
[in a new window]
|
Fig 4. Sagittal, T2-weighted, fat-saturation, MRI scan of the pelvis, showing the rectum (arrow) and bladder pushed forward.
|
|
View larger version (123K):
[in this window]
[in a new window]
|
Fig 5. Sagittal, T2-weighted, MRI scan of the thoracic spine, showing air in the epidural space (arrow).
|
|
View larger version (103K):
[in this window]
[in a new window]
|
Fig 6. Sagittal, T1-weighted, MRI scan of the thoracic spine, showing air in the epidural space (arrow).
|
|
The infant underwent surgery with a combined pediatric surgical
and neurosurgical team. The presacral collection was drained
surgically through a posterior sagittal approach. Meconium-stained
fluid was drained, and the rectum was defunctioned with a divided
sigmoid colostomy. An L5–S2 laminectomy was performed.
The epidural space contained purulent material, with striking
greenish discoloration of the epidural fat (Fig
7). There was
a separate collection in the right parasacral region, with a
possible sinus tract extending anteriorly. There was no cerebrospinal
fluid leak. Specimens for histologic examinations were taken
from the epidural fat and from the parasacral cavity and its
anterior extension.
The histologic examination of the right parasacral tissue showed
large-bowel wall composed of mucosa, submucosa, and muscularis
propria. The epidural material showed adipose tissue containing
pigment-laden macrophages and material containing deeply pigmented
hair shafts. No epithelial tissue was identified. The purulent
material from the epidural space grew
E coli and enterococcal
species. The epidural tissue showed inflammation and the presence
of meconium.
A gastrograffin enema on the 10th postoperative day showed a small fistulous tract from the upper rectum, above the levator ani muscle, extending posteriorly into the presacral space (Fig 8). The presacral space was not injected with contrast material; therefore, no communication with the spinal canal could be demonstrated.
The postoperative course was unremarkable. The infant continued
to receive intravenous antibiotic therapy and was discharged
from the hospital 3 weeks after surgery, to continue antibiotic
treatment for an additional 3 weeks.
At 6 weeks, the presacral space was explored through combined abdominoperineal incisions. Inflammatory tissue that was firmly attached to the posterior wall of the rectum, 4 cm from the anus, was removed. This tissue was confirmed, in the subsequent histologic evaluation, to be the obliterated fistula and residual cavity. The postoperative course was unremarkable. At 3 months, results of a contrast examination of the rectum were reported as normal and the divided colostomy was closed.
|
DISCUSSION
|
|---|
Neuroenteric fistulas are rare congenital malformations thought
to result from persistent communication between the embryonic
neural tube and gut tissues.
2 Isolated neuroenteric fistulas,
without involvement of bone or soft tissues, are even rarer.
3 By definition, a fistula is a communication between 2 epithelial
surfaces. It may not involve the bone but it should therefore
go through the soft tissues to reach the epidural fat. The biopsy
of the right parasacral mass showed epithelial tissue and bowel
muscle reaching the epidural space.
Embryologically, the spinal cord is formed from 2 parts. The segment from the medulla to the mid-lumbar region is formed by primary neurulation. The more distal segment is formed by a process called "canalization and retrogressive differentiation."4 The notochord arises from the notochord process on day 15 of embryonic life. It causes the overlying ectoderm to differentiate into the neural plate, the lateral parts of which thicken to form the neural folds, which eventually close to form the neural tube (neurulation) on day 17. When neurulation is complete, the neural tube separates from the overlying ectoderm. The posterior neuropore (caudal end) closes at 
27 days. Canalization is the process by which the neural tube elongates caudal to the posterior neuropore. During this process, the notochord and neural epithelium fuse, forming a caudal cell mass. Canalization of this cell mass and cellular necrosis of parts of it (retrogressive differentiation) lead eventually to formation of the conus medullaris and filum terminale. At 25 days, the notochord separates from the gut and the neural tube and, with the surrounding mesenchymal cells, forms the somites. Each somite has a medial sclerotome, which eventually forms the vertebral bodies, and a lateral myotome, which forms the paraspinal muscles. Notochordal remnants between the vertebrae become the nucleus pulposus of the intervertebral disk.5
A persistent connection between the endoderm and ectoderm is thought to result in splitting of the notochord. In the most severe form of split-notochord syndrome, communication exists from the intestinal cavity through the spinal canal and into the skin of the back. Any part of that fistula may become obliterated, leaving less complete fistulas and cysts. A persistent distal end of this fistula forms an enteric sinus, a persistent middle portion forms a neuroenteric cyst, and a persistent proximal end may result in a neuroenteric fistula4 (Fig 9). Feller and Sternberg described the split-notochord syndrome, which is characterized by the triad of vertebral anomalies, a posterior mediastinal mass, and intestinal duplication.6 The majority of neuroenteric fistulas occur in the thoracic region and involve the foregut or the midgut.7 Rarely, they can arise in the lumbosacral region and involve the hindgut.2,8,9
In our case, the presacral cyst clearly communicated with the
rectum; that communication was demonstrated with the gastrograffin
enema. The failure to demonstrate communication between the
presacral cyst and the epidural space could have been the result
of insufficient pressure within the presacral cyst, because
the contrast material was injected into the rectum and not into
the cyst. The striking meconium staining of the epidural fat,
the mixed intestinal flora within the epidural space, the pigment-laden
macrophages, and the demonstration of colonic mucosa and submucosa
in the tissue taken from the right parasacral region are strong
indicators of the existence of such fistulous communication.
The hair that was demonstrated in the epidural tissue was most
likely swallowed with amniotic fluid during embryonic life and
reached the epidural space through the neuroenteric communication.
However, teratomas within fistulas have also been described.
10 The air in the epidural space could have been intestinal gas
entering through the fistulous communication, although gas produced
through bacterial metabolism cannot be excluded as an alternative
cause.
We were surprised at the early colonization of this breastfed infant, with extensive deep infection with both enterococci and E coli by 20 hours of life. Although this is unusual for Western countries, it would be a common observation in rural communities such as those in Pakistan.11
SEAs are uncommon among children12,13 and, to our knowledge, a SEA has not been observed previously in a newborn infant. Nejat et al1 described a thoracic epidural abscess in a 20-day-old infant. In the series reported by Auletta and John,13 the youngest patient was 5 weeks of age and the average age was 8 years. None of their 8 cases or the other 26 cases they extracted from the literature had a recognized associated neuroenteric cyst, although the 5-week-old infant was described as having the Currarino triad (anorectal malformation, sacral bony anomaly, and presacral mass).12 Many of those cases were reported before the advent of MRI, which has allowed better recognition of associated anomalies.
The predominant organism in SEAs in most series is Staphylococcus aureus, consistent with a different route of infection. The diagnosis of SEA should trigger a search for the underlying cause, and the presence of coliforms should suggest the possibility of a connection to the colon.
Early diagnosis and aggressive treatment of this condition are essential, because delay can lead to catastrophic results, including sepsis, spinal deformity, permanent neurologic deficits, and death.12,14–16 A diverting colostomy is an important part of the treatment of children with neuroenteric fistulas. This prevents contamination of neural tissues and enables the physician to control the infection while awaiting definitive surgical treatment of the fistula.
|
FOOTNOTES
|
|---|
Accepted May 12, 2004.
Address correspondence to Thorsten V. Stanley, FRCP, Department of Paediatrics, Wellington School of Medicine and Health Sciences, PO Box 7343, Wellington, New Zealand. E-mail: paedtvs{at}wnmeds.ac.nz
|
REFERENCES
|
|---|
- Nejat F, Ardakani SB, Khotaei GT, Roodsari NN. Spinal epidural abscess in a neonate.
Pediatr Infect Dis J. 2002;21
:797
–798[Medline]
- Durham MM, Chahine AA, Ricketts RR. Presacral neuroenteric fistula presenting with meningitis and vaginal fistula: a case report.
J Pediatr Surg. 1998;33
:1558
–1560[Medline]
- Rivierez M, Buisson G, Kujas M, et al. Intramedullary neurenteric cyst without any associated malformation: one case evaluated by MRI and electron microscopic study.
Acta Neurochir. 1997;139
:887
–890
- Barkovich AJ.
Neuroimaging. 2nd ed. New York, NY: Raven Press; 1995:477–540
- Skandalakis JE, Gray SW.
Embryology for Surgeons. Baltimore, MD: Williams & Wilkins; 1994: 88–92, 248–253
- Feller AH, Sternberg H. The notochord and its genesis [in German].
Virchows Arch. 1929;272
:613
–640
- Agnoli AL, Laun A, Schonmayr R. Enterogenous intraspinal cysts.
J Neurosurg. 1984;61
:834
–840[Medline]
- Alrabeeah A, Gillis DA, Giacomantonio M, Lau H. Neuroenteric cysts: a spectrum.
J Pediatr Surg. 1988;23
:752
–754[CrossRef][Medline]
- Gupta DK, Deodhar MC. Split notochord syndrome presenting with meningomyelocele and dorsal enteric fistula.
J Pediatr Surg. 1987;22
:282
–283
- Lynn AM, Stuntz JT, Ward BH. Pneumocephalus: an unusual presentation of a persistent neurenteric fistula.
J Pediatr. 1978;98
:818
–820
- Adlerberth I, Carlsson B, De Man P, et al. Intestinal colonization with enterobacteriaceae in Pakistani and Swedish hospital-delivered infants.
Acta Paediatr Scand. 1991;80
:602
–610[Medline]
- Liu KA, Luhmann JD. Spinal epidural abscess in preverbal children: a case report with Currarino triad.
Pediatrics. 1999;104
:1139
–1142[Abstract/Free Full Text]
- Auletta JJ, John CC. Spinal epidural abscesses in children: a 15-year experience and review of the literature.
Clin Infect Dis. 2001;32
:9
–16[CrossRef][Web of Science][Medline]
- Enberg RN, Kaplan RJ. Spinal epidural abscess in children: early diagnosis and immediate surgical drainage is essential to forestall paralysis.
Clin Pediatr. 1974;13
:247
–253
- Walter R, King J, Manley J, Rigamonti D. Spinal epidural abscess in infancy: successful percutaneous drainage in a nine-month-old and review of the literature.
Pediatr Infect Dis J. 1991;19
:860
–864
- Agrawal D, Suri A, Mahapatra AK, Sharma MC. Intramedullary neurenteric cyst presenting as infantile paraplegia: a case and review.
Pediatr Neurosurg. 2002;37
:93
–96[Medline]
PEDIATRICS (ISSN 1098-4275). ©2004 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
TOP
ABSTRACT
CASE REPORT
