Eur Spine J
DOI 10.1007/s00586-014-3218-x
CASE REPORT
Diastematomyelia in congenital scoliosis: a report of two cases
Stefan Gavriliu • Costel Vlad • Ileana Georgescu •
Gheorghe Burnei
Received: 19 June 2013 / Revised: 23 January 2014 / Accepted: 25 January 2014
Ó Springer-Verlag Berlin Heidelberg 2014
Abstract
Introduction Even if diastematomyelia is a rare condi-
tion, it always has to be identified in case of diagnosing and
treating a case of congenital scoliosis. The consequence of
missing the diagnosis of such a malformation may be
devastating to the involved patient. This paper wants once
again to make aware the physicians of the eventual pre-
sence of a spinal dysraphic malformation when dealing
with a congenital spinal malformation.
Methods The presence of diastematomyelia was noticed
in two of our cases, one with congenital scoliosis and
another with congenital kyphoscoliosis. First of all in these
cases, we performed the resection of the bony septum,
followed by spinal fusion in a single-stage surgical pro-
cedure. We noticed no complications during and after
surgery related to the resection of the bony septum.
Results In both patients, we obtained partial correction
and stabilization of the congenital spinal malformation
after a safe excision of the bony septum.
Conclusions Diastematomyelia is a rare condition. It has
to be taken into consideration when dealing with a con-
genital scoliosis. The first step in the surgical procedure has
to be the resection of the diastematomyelic septum. In case
of a scoliosis ranging up to 30° and not presenting a pro-
gressive potential, the expectative-evaluation attitude is a
correct one.
S. Gavriliu (&)
C. Vlad
I. Georgescu
G. Burnei
Clinical Emergency Hospital for Children ‘‘Maria Sklodowska
Curie’’, 20 C-tin Brancoveanu, Bucharest, Romania
e-mail: banteo@gmail.com
S. Gavriliu
G. Burnei
‘‘Carol Davila’’ University of Medicine and Pharmacology,
Bucharest, Romania
Keywords Diastematomyelia
Congenital scoliosis

Single-stage surgery
Progressive potential
Introduction
Diastematomyelia is a rare malformation in which the
spinal cord is divided in the sagittal plane on a variable-
length segment. This malformation is due to the presence
of a transfixiant septum of bony, cartilage or fibrous origin
from the vertebral body, with an intracanalar trajectory
from anterior to posterior. In the literature, there remains
confusion in the definition of this malformation, which has
often overlapped as interpretation with diplomyelia (true
duplication of the spinal cord). Pang et al. [1] proposed to
introduce both types of malformation in a complex dys-
raphic status because symptoms and associated malfor-
mations are common.
Historically, the first to introduce the term of diaste-
matomyelia is Oliver (1837), who made the first extensive
description of such a case. The oldest known case in lit-
erature is a specimen discovered in Israel, dating from 100
AD, which presented a diastematomyelic septum, in the
presence of a congenital scoliosis due to a segmented
hemivertebra [2].
In terms of etiology, the causes of this malformation are
still unknown. It is assumed that the malformation may be
triggered in the embryogenesis by the existence of an
accessory neuroenteric ductus or the persistence of the
neuroenteric ductus on a certain spinal area, which would
explain the relatively frequent association of tumor
development in the affected region (Brenner 1952).
Topographically speaking, predominant localization in
over 75 % of the cases is distal to T7 down to S1. Exceptional
location may be noticed in the cervical region [3].
123

Fig. 1 X-rays showing the presence of a congenital scoliosis with
T12 left hemivertebra and an L1–L3 rahischisis
In terms of imaging, plain radiography is not relevant.
The presence of associated malformations such as con-
genital scoliosis with hemivertebrae or butterfly vertebrae
or rachischisis raises the suspicion of the existence of a
dysraphic status of the spinal cord. Therefore, it is abso-
lutely necessary to perform CT and MRI to reveal occult
malformation in all ‘‘X-rays undetectable’’ cases. The
presence of a septum of cartilage or fibrous nature will be
obvious at MRI examination. It is imperative to carry out
these imaging investigations before starting the treatment
of congenital scoliosis, because the presence of such dys-
raphic defects may have adverse consequences in an
attempt to redress the scoliosis curve (paralysis, inconti-
nence, etc.) [4]. Following, the rule is to address the dia-
stematomyelia first and then the congenital scoliosis.
Cases report
We treated surgically 37 cases of congenital scoliosis by
different types of approaches and instrumentations in
2004–2010 [5]. We noticed two cases associated with the
presence of diastematomyelia. The most frequent associa-
tion is represented by diastematomyelia and hemivertebra.
In these cases, surgery may consist in bony septum removal
and hemivertebra removal by a single posterior approach or
a combined one, anterior and posterior with hemivertebra
resection. Some authors state the fact that a single posterior
approach has few advantages like a reduced number of
incisions, less operating time and recovery time. The dis-
advantages of this single-stage intervention are related to a
reduced visualization in the operating field, increasing this
123
Eur Spine J
way the neurological risks, more probability of retraction
of the dura and nerve roots and increased blood loss [6].
When operated by a skilled surgeon, the single-stage
intervention by a posterior approach only tends to offer a
better correction of the scoliosis with less neurological
complications [7, 8].
Case I is represented by a 7-year-old girl with poor
social status, the diagnosis of congenital scoliosis with a
left T12 hemivertebra, L1–L3 spina bifida associated with
the presence of an L2–L3 diastematomyelic bony septum
(Figs. 1, 2). The patient had clinical neurological phe-
nomena, such as pelvic limb paresthesias, a more pro-
nounced hypotrophy in the calf, muscular weakness and
diminished Achilles tendon and patellar reflexes. We
noticed a limb length discrepancy with a shortening of
1 cm of the right lower limb, too.
The presence of these clinical manifestations required
surgery that consisted in:
1. Excision of L2–L3 diastematomyelic septum;
2. Left T12 hemivertebra resection;
3. Correction and stabilization of the scoliotic curve by
T11–L3 somatic synthesis (Fig. 3).
Postoperative evolution was favorable, with the disap-
pearance of neurological signs, but at 3 years postopera-
tively we found that the lower limb hypotrophy was
persistent and the lower limb length discrepancy increased
to 4 cm (Fig. 4). Next, we intend to perform a transient
epiphysiodesis to equalize the lower limbs length.
Case II is represented by a 12-year-old boy with a
congenital thoraco-lumbar kyphoscoliosis due to mixed
formation and segmentation defects (T11–T12 anterior
spinal block, T11–L1 butterfly hemivertebrae) associated
with the presence of a T12 diastematomyelic septum
(Fig. 5). MRI examination revealed an association with an
about 2.5 cm syringomyelic cavity located distal to the
diastematomyelic septum, where the two medullar seg-
ments met again. Surgery consisted in:
1. Resection of the diastematomyelic septum;
2. Posterior in situ fusion by rib bone allografts (Fig. 6).
Postoperative evolution was a favorable one. The sco-
liotic curve remained constant and the kyphosis showed
partial spontaneous correction after an effective posterior
fusion after a 3-year follow-up (Fig. 7).
Discussions
Even if rare, this malformation has always to be considered
in the presence of spinal congenital anomalies, especially
congenital scoliosis, even if the patient does not express
any clinical manifestation. Therefore, it is absolutely
Eur Spine J
Fig. 2 CT exam showed the presence of a diastematomyelic septum located at L2–L3, and an abnormality of shape and position of the vertebral
arch
Fig. 3 Postoperative image after the septum excision and T12
hemivertebra resection. T11–L3 somatic synthesis by a transpleural-
retroperitoneal Hodgson approach
required to perform all imaging investigations necessary to
establish a precise diagnosis in the presence of vertebral
column malformations (plain radiography, CT, CT-3D and
MRI) before the correction of the spinal axial deviation to
avoid neurological complications of those induced by the
presence of a diastematomyelic septum.
The presence of clinical signs is an absolute indication for
surgical treatment. Absence of clinical signs is a dilemma for
Fig. 4 Clinical image 3 years
after surgery. A 4 cm limb
length discrepancy was noticed
many authors in the surgical treatment initiation: expectative-
evaluation attitude or surgery? The surgeons indicating a
surgical approach to the malformation consider this attitude
due to the presence of the bony septum which fixes the
medulla. Between the bony growth of the spine and spinal
cord development, there is a gap, so that spinal traction is a
phenomenon that will prevent upward migration. The tension
upon the medulla will increase with growth which may trigger
a hypovascularized marrow and a progressive neurological
123
 Eur Spine J

Fig. 5 X-rays and CT-3D images. We noticed the presence of a congenital kyphoscoliosis and a T12 diastematomyelic septum

Fig. 6 During surgery, the diastematomyelic septum was removed and T9–L3 posterior in situ fusion with rib allografts was performed

Fig. 7 Clinical and X-rays exams 3 years after surgery showed a stabile scoliotic curve and a spontaneous correction of the kyphotic curve. The
patient presented no altering in the clinical status after the removal of the diastematomyelic septum

deficit, sometimes to complete paralysis of lower limbs. The therapeutic algorithm should be a single-stage sur-
That’s why it would be justified to take surgical attitude if gery addressed first to the diastematomyelic septum and
diastematomyelia is diagnosed, even if asymptomatic [9]. next to the scoliotic curve correction if it reaches a 30°–40°

123
Eur Spine J
Cobb angle. This attitude is even more effective as the
progressive potential is higher. The potential of increasing
the curve is related to the asymmetry in growth potential on
the convex and concave sides of the curve [10]. Not
respecting these steps may cause irreversible and irrepa-
rable damages.
It is known that about 25 % of congenital scoliosis cases
have no progressive potential and do not require surgery.
Our indication for such patients, if they associate the pre-
sence of diastematomyelia, is the expectative-evaluation
attitude by clinical and radiological evaluation. If we face a
congenital scoliosis without indication for surgery, but
there is evidence of dynamic growth in size of the diaste-
matomyelic septum with risk of occurrence of phenomena
of spinal canal stenosis, our opinion shifts to surgery by
excision of the diastematomyelic septum, without waiting
for possible clinical manifestations of the malformation.
If case I would have been diagnosed before the age of
3 years, when the first spurt of growth of the child instal-
led, and would have undergone surgery, it could have been
possible that the difference in length of lower limbs did not
appear, although one cannot exclude the coexistence of
idiopathic limb length discrepancy or a disrupted primary
neurulation with its consequences [11]. Lately, there
appeared the notion of occult tethered spine in the literature
[12], which is the possible cause of the limb length dis-
crepancy and neurological manifestations, with no other
symptoms or signs, in this patient. Surgery was a type of
prophylaxis in case II, operated on the eve of the second
spurt of rapid growth, which, in addition to the presence of
congenital kyphoscoliosis, associated with syringomyelic
cavities. It was, therefore, considered the risk of possible
rupture of the syringomyelic cavity in evolution and that a
posterior arthrodesis in the presence of a congenital
kyphoscoliosis would spontaneously correct the kyphotic
curve during growth.
Conclusions
This spinal malformation, although rare, must be properly
and timely diagnosed by imaging with aid of the entire
arsenal of means to avoid devastating and irreversible
complications. Asymptomatic cases associated with
congenital scoliosis without surgical indication should be
carefully monitored. The most sensitive sign of developing
any complications requires surgery. Compliance with
treatment algorithm that requires mandatory initial
approach of diastematomyelia and next the congenital
scoliosis ensures safety to the patient and surgeon.
Acknowledgments The senior author GB operated all the cases
mentioned in our clinics experience.
Conflict of interest None of the authors has any potential conflict of
interest.
References
1. Pang D, Dias MS, Ahab-Barmada M (1992) Split cord malfor-
mation: part I: a unified theory of embryogenesis for double
spinal cord malformations. Neurosurgery 31:451–480
2. Edelson JD, Nathan H, Arensburg B (1987) Diastematomyelia––
the ‘‘Double-barreled’’ spine. J Bone Joint Surg Br 69(8):188
3. Ohwada T, Okada K, Hayashi H (1989) Thoracic myelopathy
caused by cervicothoracic diastematomyelia. J Bone Joint Surg
Am 71(2):296–299
4. Akbarnia BA, Yazici M, Thompson GH (eds) (2011) The
growing spine. Management of spinal disorders in young chil-
dren. Springer-Verlag, Berlin, Heidelberg, p 67
5. Burnei G, Georgescu I, Gavriliu S, Vlad C, Dan D (2007) The
correction of congenital scoliosis by somatic instrumentation
associated to hemivertebral resection and discectomy. In: World
IV interdisciplinary congress on spine care, July 29th–August 1st,
2007, Istanbul, Turkey
6. Zhang J, Shengru W, Qiu G, Yu B, Yipeng W, Luk KD (2011)
The efficacy and complication of posterior hemivertebra resec-
tion. Eur Spine J 20(10):1692–1702
7. Obeid I, Bourghli A, Vital JM (2012) Lumbar hemivertebra
resection by posterior approach for congenital scoliosis. Eur
Spine J 21(12):2721–2723
8. Obeid I, Bourghli A, Vital JM (2013) Thoracic hemivertebra
resection by posterior approach for congenital scoliosis. Eur
Spine J 22(3):678–680
9. Miller A, Guille JT, Bowen JR (1993) Evaluation and treatment
of diastematomyelia. J Bone Joint Surg Am 75(9):1308–1317
10. Arlet V, Odent T, Aebi M (2003) Congenital scoliosis. Eur Spine
J 12:456–463. doi:10.1007/s00586-003-0555-6
11. Kaufman BA (2004) Neural tube defects. Pediatr Clin North Am
51:389–419
12. Tu A, Steinbok P (2013) Child’s nervous system. 29:1635–1640.
doi:10.1007/s00381-013-2129-1
123