The range of conus levels,
in
brackets, and the mean ± 1
standard deviation,
in
parentheses, by age intervals rounded
off to the nearest year for
group-11
subjects (patients with
symptoms but whose
MR
findings were normal) were:
0-2
years [T12-L2.5](13 ± 1.45),
3-4
years [T12.5-L2.5](13.5 ±
1
.3
0)
,
5-6
years [T12.5-L2](13.2 ± 0.97),
7-8
years
[T12-
L2](14.1 ± 1.46),
9-10
years [T12-L2.5](13.2 ± 1 .48),
11-
12
years [L1-L2](12.9 ± 0.78),
13-14
years
[T12-L2
.5](12.8
± 1.48),
15-16
years [L1-L2.5](12.0 ± 1.73),
17-18
years
[T12.5-L2](13.0
± 1.22),
19-20
years [L
1-L
1.5](13.7 ±
0.
58).
The mean conus level for groups I and
II
was compared
and
found to
be
the same at each age interval at a confidence
level
of 95
%.
Consequently, groups I and
II
were compared
as
a whole. The means and standard deviations were 13
.5
±
0.85 and 13.2 ± 1.32 for groups I and
II
, respectively. There
was no statistical difference between these groups at a
confidence level of 95
%.
These two groups were therefore
combined, giving a grand mean and standard deviation for
both groups of 13.3
± 1.13. This corresponds to
an
approx-
imate anatomic level of the L
1-L2
intervertebral disk space
and
two standard deviations would
be
T12 to
L2-L3
. The
ranges and averages for the combined groups are given
in
Figure 2.
The conus level
in
the 1 00 normal adults ranged from
T11-
T12 to
L2-L3
with the average level at L
1.
The patients with surgically proved tethered cords had
conus levels ranging from
L3
to
S4
, with the average level of
termination at
L4-L5.
No age-related pattern was observed
in
these patients.
There were radiographs of the lumbosacral spine available
for review on 96/184 (52%) of the group
I and group
II
subjects.
We
found eight children with transitional vertebrae,
two children with four lumbar vertebral bodies, and
two
children with six lumbar vertebral bodies. The appropriate
corrections were made
in
each instance to reflect the true
level
of conus termination. The conus level of termination was
corroborated by myelography
in
six cases from group I and
in
five cases from group
II.
Discussion
The volume of data used to develop the present-day criteria
for the normal conus level
in
children
is
in
fact rather small.
V>
·
.::
""
T11
::l
'0
There are three anatomic studies of children
in
the recent
literature that evaluated conus level by the dissection of
cadavers. The most widely quoted study,
by
Barson [3]
in
1969,
is
actually a study of cord levels
in
the fetus (only 12
specimens over the age of 1 year were examined).
He
dem-
onstrated conclusively that the conus level does ascend
throughout fetal life, but, on the basis of data from a very
small number of infants and children, suggested that the adult
level of L
1-L2
was attained
by
the age of 8 weeks.
In
another
anatomic study, Jit and Charnalia [ 4] reported the average
location of the tip of the conus
in
1 0 newborn infants
as
the
middle of
L2
, with a range from L
1-L2
to
L2-L3
. No older
children were studied. These researchers accepted the adult
level
as
L
1-L2
and noted that there was a change of only half
a vertebral body from birth to adulthood. The other anatomic
study with information pertinent to this problem was published
in
1972
by
James and Lassman [5], who examined the conus
level
in
autopsies of 25 children ranging
in
age from newborn
to 8 years. They reported that the conus reached the
L2
level
no
later than the age of 5 months, but they examined only
three specimens over the age of
1 year.
The combined anatomic data from these three studies
contain or.ly 15 measurements
in
children over the age of 1
year. This
is
insufficient evidence to justify the accepted level
of termination of the normal conus throughout childhood.
No
imaging studies of significant size have been performed to
better define the level of the normal conus
in
children, even
though it
is
well accepted that myelography
[1
, 6] and
MR
imaging [7]
can
accurately make this determination.
Our study contradicts the conventional wisdom that the
conus
medullaris ascends throughout childhood.
We
found
that there is
no
significant difference
in
the level of conus
termination among the various age groups of normal children
(see
Fig
.
2).
At any age from birth to 20 years, a conus level
of termination at or above
L2-L3
was normal. The subjects
with a conus level at the L3 level provided some difficulty.
We
encountered one normal 3-year-old girl with a conus at this
level
(Fig
.
3)
. There were also
two
patients with surgically
proved tethered cords that terminated at this level. This
is
not
surprising when we note that Reimann and Anson [8] deter-
mined from their study and review of
801
adult cadavers that
1.8% of normal adults
will have a conus level of termination
at L3.
--
~
----
-
t
---
-l
---
-t
-----
-----
r
---
-----------±-------
--
-
f ! -
-
"'
T1
2
:2
V>
L1
1!
0
u
L2
Fig.
2.-Average
conus
level
(dots)
and range
(vertical bars) are indicated
for
184 normal chil-
dren, ages newborn
to
20 years, and for 100
adults.
0
.
9-
0
a;
>
"'
-'
1
L3
L4
L5
[31]
0
-
[25] [23]
[14]
[12]
[11] [21]
[10]
[17]
[20]
(1
00]
2 3 4 5 6 7 8 9 10
11
12 13 14 15 16 17 18 19
20
Adul1
Ag
e in Years