Docstoc

Cupids Bow Contour ofthe Vertebral Body Evaluation

Document Sample
Cupids Bow Contour ofthe Vertebral Body Evaluation Powered By Docstoc
					Karence Elizabeth

K. Chan, MD Barrett-Connor,

#{149}

David MD

J. Sartoris,
#{149}

Debra

MD #{149} Parviz T. Trudell, RA

Haghighi,
#{149} Donald

MD Resnick,

#{149} Paul

Sledge,

PhD

MD

Cupid’s Bow Contour Evaluation ofPathogenesis and Imaging-Histopathologic
PURPOSE: bow contour
tometry and

ofthe

Vertebral Body: with Bone Densitometry Correlation’

To investigate the Cupid’s by means of bone densiimaging-histopathologic

correlation. MATERIALS AND METHODS: Radiographs and histologic slices were obtained in 64 cadaveric thoracolumbar spines to assess the morphology, distribution, and histopathologic features of the Cupid’s bow contour. Dual-energy x-ray absorptiometry and radiography were performed in the lumbar spine in 406 healthy subjects. Bone density and body height and weight were then related to the presence of the Cupid’s bow contour. The Cupid’s bow contour was cornpared with the fish vertebra of osteoporosis and Schmorl node.
RESULTS: No clinically important

relationship was found bar bone density, body weight, and prevalence
pid’s bow contour.

between lumheight and of the Cuexarni-

Histologic

nation showed thickened bone in the Cupid’s bow end plate with annular fibers inserting into this region. In cadavers, the Cupid’s bow contour occurred at multiple lumbar and thoracic levels, with the highest frequency in the lower lumbar spine. Lateral radiographs enabled better detection of the contour change. CONCLUSION: The Cupid’s deformity is a developmental nomenon that is unrelated to penia or mechanical stress on spine. bow pheosteothe

end plates have a vanety of curvatures on both frontal and lateral radiographs. Some of the contour variations are caused by diseases; others are considered normal variants. The early literature has described a parasagittal concave curvature in the inferior end plates of the vertebral bodies in the lumbar spine (Fig 1). The appearance of this curve on the anteropostenior radiograph was similar to a Cupid’s bow (1) and was named accordingly. The Cupid’s bow contour is seen on transverse computed tomographic scans as two parasagittal circular areas of low attenuation with a sclerotic rim in the posterior portion of the vertebral body (2,3). Even though the cause of the Cupid’s bow remains unknown, it has been speculated that this contour could be a variation in notochordal development (1). To our knowledge, no study has been performed to investigate whether the altered contour has any pathologic
ERTEBRAL

V

implications

such

as

an

association

with osteoporosis. The purpose of our study was to evaluate the bone density in subjects with a Cupid’s bow contour that involved at least one vertebral level by means of dual-energy x-ray absorptiometry (4,5) and to determine whether the Cupid’s bow contour is associated with osteoporosis or mechanical stress. In addition, we evaluated cadavenic thoracolumbar spines to understand

I I;
Figure
lumbar Cupid’s

‘-t,4

1.
spine bow

Anteroposterior in a 37-year-old

radiograph of the man shows the

contour at the inferior of L-2 to L-4. The contour appears sagittal concave curvature.

end plates as a para-

the the

morphology Cupid’s bow

and distribution contour to deter-

of

Index
ties,

terms: 32.1499,

Bones, 33.1499
1997;

mineralization,
#{149} Spine,

32.56,

33.56
defect,

#{149} Osteoporosis, 32.1499, 33.1499

32.56,

33.56

#{149} Spine,

ahnormali-

developmental

Radiology

202:253-256

From

the

Departments

of Radiology

(K.K.C.,

D.J.S.,

DR.),

Pathology

(PH.),

Family

Preventive

Medicine cal Center,

(P.S.), and Medicine (E.B.C.), School of Medicine, San Diego; and the Departments of Radiology
Affairs Medical Center, 3350 La Jolla Village

University of California, San Diego Mcdi(K.K.C., D.J.S., D.T.T., DR.) and Pathology
Dr. San Diego, CA 92161. From the 1995

(PH.),

Veterans

RSNA scientific assembly. Received April 8, 1996; revision 15; accepted July 18. Supported in part by U.S. Department dress reprint requests to DR. . RSNA, 1997

requested June 3; revision of Veterans Affairs grant

received July no. SA-360. Ad-

253

a. Figure

2.

Definition

of distinct
Each vertebral

b. vertebral
diagram end

end-plate

variations.
variation vertebra

Diagram

c. shows

the morphologic

criteria

used

d. to define

specific

types

of verte-

bral end-plate variations. posterior portion of the

shows vertebral plate. (b) The fish

in the lateral is a symmetric

view. (a) The Cupid’s bow contour biconcavity that extends throughout

is a smooth the end

concavity at the plate. (c) Verte-

bral collapse

is

angular

deformity

involving

the entire

end

plate.

(d) A Schmorl

node

is an irregular

focal

indentation

of the end

plate.

T-1O

.

Ti 1 T12 Li

L-5 5
Total Number
10 15 20 25 Bow

of Specmens

W’rh

Cuprcbs

Contour

Figure

3. the Cupid’s

Graph bow

shows contour

the distribution of from T-iO to L-5.
number of specibow is located in black shading

Gray mens

the represents the number of specimens with a Cupid’s bow in the inferior end plate. The
most plates frequent locations of L-2 to L-4. are the inferior end

shading indicates the in which the Cupid’s superior end plate, and

mine
Cupid’s vertebral vertebrae

a clear

distinction

between

the
in fish (7). a. Figure 4. Anteroposterior degrees of parasagittal
racic vertebral body radiographs of vertebral

bow and other alterations body shape, including (6) and Schmorl nodes

MATERIALS Cadavers

AND

METHODS

b.
bodies from elderly men show various

concavities
(a) and each

of the Cupid’s
lumbar vertebral

bow
body

in the inferior
(b).

end

plate

of each

tho-

Radiographs were obtained in the anteropostenor and lateral views from 64 cadaveric thoracolumbar spines. The spines were then sectioned in the sagittal

eric

thoracolumbar was

plane were

into then

3-mm-thick obtained
of the

slices. Radiographs of the thin slices. The
Cupid’s bow contour

tenor and lateral
contour
was sectioned

spine in the anteroposviews. A Cupid’s bow identified. The spine then
into thin coronal slices, and

were
x-ray

obtained
absorptiometry

by means

of dual-energy
Hologic,

(QDR-2000;

morphology

was characterized from the T-iO to L-5 vertebral levels. The Cupid’s bow contour was defined on the lateral view as a smooth concave curvature with its center located at the posterior portion of the vertebral end plate (Fig 2a). The curvature had a
steeper slope posteriorly than anteriorly.

the morphology of the Cupid’s bow was further characterized. Histologic slides were then prepared and evaluated.

Boston, Mass). Measurement of bone mmeral density enabled us to determine whether the appearance of a Cupid’s bow contour was associated with the prevalence and severity of osteoporosis.

Subjects
After the Cupid’s fined, the bone density spine was evaluated subjects (age range, years). The subjects longitudinal study of Bernardo Heart and contour was deof the lumbar in 406 healthy elderly 53-98 years; mean, 76 are participating in a bow

RESULTS Cadavers
Of the 64 thoracolumbar spines, one had an incomplete L-4 vertebral body and 13 had an incomplete L-5

The superior and inferior end plates of each vertebral body were evaluated for the presence of the Cupid’s bow contour. The appearances of fish vertebra, vertebral
collapse, and The Schmorl fish node were end also plate noted

normal
Chronic

aging
Disease

(Rancho
sur-

and
contour.

were

compared

with
vertebral

the Cupid’s

bow
was

vertebral proper
topsy.

body removal

resulting of the

from imspine at au-

defined as a smooth concavity that extended throughout the entire vertebral end plate (Fig 2b). Vertebral collapse was defined as angular deformity of the vertebral end plate (Fig 2c). A Schmorl node was defined as an irregular focal indentation of the vertebral end plate (Fig 2d).
Radiographs were obtained of a cadav-

vey) funded by the National Institutes of Health. Analysis of the lateral radiographs enabled classification of the subjects as
either contour those with or those at least without one Cupid’s a Cupid’s bow bow

contour. Body height and weight were recorded for each subject, and bone mmeral densities for the whole body and the
spine (frontal and lateral projections) (8)

Thirty-four spines contained at least one vertebral body with a Cupid’s bow contour, whereas in 30 spines, a Cupid’s bow contour was not apparent at any level. Ten spines had fish vertebrae, nine had collapsed vertebrae involving at least one spinal

level,

and

10 had

a Schmonl

node
January

at 1997

254

#{149} Radiology

.:

:i.

Tb ‘111

Tb Til
-

T.12
Li

1-12
Li ..-_ .-

_________
#{149}
-

L-2

_______j

L3
L4 L5 fl

1
----

L.3
L4

-

]
2 4

L5

-

.1
is o

0

6
of

e Bow Contours

io

2

o

s

ro

25

o

Number

Cupid’s

Number

of Cupids

Bow Contours

.,,,-

a. b. Figure 6. Graphs show (a) the distribution of Cupid’s bow contour of superior end plates both anteroposterior (black shading) and lateral (gray shading) views and (b) the distribution of Cupid’s bow contour of inferior end plates in anteroposterior (black shading) and lateral
(gray
shading) views.

in

Lateral

views

enable

better

detection

of the

Cupid’s

bow

contour

for

both

superior

and

inferior

end

plates.

without intervening sue. Degenerative jacent to the region

cartilaginous disk alterations of the Cupid’s

tisad-

bow found

were

less

frequently
cell clones

found.
were

No the
in

notochordal-type

in the disk

directly

beneath

Cupid’s bow. Groups of chondroid cells were noted within the disk

the midline.

Subjects
Lateral radiographs of the lumbar

spine from the 406 subjects showed that 118 had at least one Cupid’s bow contour and 288 had no contour. Corn-

r-%
Figure
graph edge 7. of the Low-magnification
section of a coronal

Cupid’s

bow

photomicroobtained at the (cli) shows thicken-

panson of the two groups demonstrated no difference in body height and weight (Fig 8). In addition, no substantial difference was found in bone mineral density between the two groups (Fig 9). No increased prevalence or severity of osteoporosis was

found
more

in subjects
Cupid’s bow

who

had

one

or

contours.

I
radiograph
an elderly indicative
each

ing of the bone

plate

with

calcium

deposi-

of the lumbar man shows of the Cuend plate.

tion. Note that the collagen rectly into the bone plate. Cu id’s bow however a (c) lies between the bone collagen fibers insert into
the gradual transition

fibers insert diAdjacent to the la er of cartila e plate and disk. The the cartilage. Note

.

inferior

(0 between
stain;

the two
original

areas. (Hernatoxylin-eosin magnification, x4.)

Although the Cuwas seen in both superior end plates, observed more
inferior end plate.

contour was noted to the L-5 level, frequency occurring spine (Fig 3). concavity on antenoviews varied. revealed shalconcavities, whereas concavities (Fig the morpholwas relatively

constant (Fig 5). The lateral view enabled better detection of the Cupid’s bow contour (Fig 6). In the 64 cadavens, 162 vertebral bodies revealed the Cupid’s bow contour in the lateral view, whereas 115 revealed the Cupid’s bow contour in the frontal view. Histologic examination (Fig 7) demonstrated a thickened bone plate and the absence of cartilaginous end-plate tissue in the region of the Cupid’s bow. The annular fibers appeared to insert into the irregularly calcified bands between the disk and the bone plate

Formation of the Cupid s bow has previously been related to notochordal regression because of its posterior location (1). On the basis of our results, however, this explanation appears to be unsatisfactory. According to Sch.morl and Junghanns (9), in prechondral and chondral phases of growth, the notochord is located in the midline as a rod passing through the cartilaginous analog of the vertebral bodies and intervertebral disks. During subsequent development of the spine, the cartilage ossifies and forms the vertebral bodies, whereas the notochord completely regresses from the vertebral border. In the intervertebral disk, the notochord is encroached on and balloons to produce intervertebra] swelling. The disk progressively differentiates into four parts: the

Volume

202

#{149} Number

I

Radiology

#{149} 255

.2

16’

16O

12
0

-120

E
U

08

0)
5,

;;;
80

.cci

‘-

-

40

::
Figure
9.
ofsubjects

---V

Comparison
who

of bone

mineral

den-

___L o Figure
Cupid’s

sities

8.
bow

Comparison
contours

of height
in the lumbar

and

weight
spine
(ti

of subjects
= 288;

without

bow

contour

had atleast one Cupid’s in the lumbar spine (gray shadthe contour (black differences in bone found between the

and
groups.

those

with
No

at least
substantial

one Cupid’s
differences

bow
were

contour
found

shading) (n = 118; gray

black

shading).

between

the two

ing) and those without shading). No substantial mineral densities were groups.

nucleus

pulposus,

the inferior

anulus

fibrosus,

and
superior apposing

two

cartilaginous
and vertebral

plates
surfaces

on
of

the
the

bodies. Focal regions that contain no cartilage exist in the cartilaginous end plates. These bare areas are replaced by a layer of calcium. The importance of this phenomenon is uncertain, but it has been proposed that these areas serve as nutritional canals (9). The results of our histologic study indicated that the thin layer of cartilage in the vertebral end plate is absent in the parasagittal region. A layer of calcium was found within this area. The cartilaginous plate is the foundation for the growing vertebral body because it contains the endochondral growth layer of the vertebral ossification center (9). We hypothesize that, owing to the absence of the cartilaginous plate in the parasagittal portions of the diskovertebral junction, the bone growth of the vertebral body is hindered at this particular region, resuiting in the appearance of Cupid’s bow concavities. Future testing of this hypothesis may necessitate examination of vertebral end plates from additional cadavers.
Findings from comparison of sub-

that the Cupid’s bow contour occurs at all levels of the lumbar spine and the lower thoracic spine and that it is present in both superior and inferior end plates. The degree of concavity
varies from one

plate as seen eral projections.
A Schmorl

on both
node

frontal

and

latby

is characterized

displacement

of a portion

of the

inter-

individual

to another

jects with and those without the Cupid’s bow contour indicated that there is no difference in body height, weight, body mass index, or bone mineral density between the two
groups. These findings further sup-

port the hypothesis bow is a developmental without pathologic In contrast to the
tion that of the Cupid’s

the Cupid’s variation implications. common concepbow is restricted

that

to the
sis

lower
cadavenic

lumbar

spine,

our
revealed

analy-

specimens

and as a function of the vertebral level. This variation may depend in part on the presence of the cartilaginous plate and calcium deposition. Our results indicate that it may be difficult to identify the Cupid’s bow contour in the frontal view. This difficulty appears to result from variations in the degree of concavity of the vertebral body. In addition, evaluation of the L-5 vertebra can be compromised because it is tilted posteriorly as part of the normal lumbar curvature. For these reasons, the lateral view enables better detection of Cupid’s bow contours. Although the degree of concavity varies, the configuration of the Cupid’s bow is relatively constant and is characterized by a concavity located in the posterior third of the vertebral body with a relatively smooth curvatune, which has a shallow angle anteniorly and a steep curvature posterionly. Our results suggest that the Cupid’s bow contour is a developmental phenomenon. Therefore, it is important to distinguish it from other vertebral end-plate deformities such as fish vertebra, vertebral collapse, and Schmonl node. Fish vertebra and vertebral collapse are associated with disorders in which there is diffuse bone weakening. These deformities are common in the lower thoracic and upper lumbar spine (6,10), where the Cupid’s bow contour also is frequently seen. Unlike the Cupid’s bow, however, the concavity of the fish vertebral end plate is centered in the middle of the end plate and affects the entire end

vertebral disk into the vertebral body due to disruption of the cartilaginous end plate (7). The Schmorl node forms depression, leading to discontinuity and an irregular contour of the verte-

a

bral end plate, and can portion of the vertebral References
1.

involve any end plate.

a

2.

Dietz GW, Christensen EE. Normal “Cupid’s bow” contour of the lower lumbar vertebrae. Radiology 1976; 121:577-579. Firooznia H, Tyler I, Golimbu C, Rafli M. Computerized tomography of the Cupid’s

bow contour of the lumbar Radiol 1983; 7:347-350.
3. Ramirez “Cupid’s

spine.

Comput

HJ, Navarro JE, Bennett WF. bow” contour of the lumbar

ver-

tebral

endplates

detected
Assist

by computed
Tomogr

to1984;

mography.

J Comput

8:121-124.
4.
Sartoris

DJ, Moscona A, Resnick D. Progress in radiology: dual-energy radiographic abfor bone densitometry-cur-

sorptiometry

5.

rent status and perspective. Ann N Y Acad Sci 1990; 592:307-325. [Discussion: 334345.] Adams JE. Osteoporosis and bone mm-

eral densitometry.
6.

Curr Opin

Radiol
Arthritis

1992;

7.

4:11-20. Resnick DL. Fish vertebrae. Rheum 1982; 25:1073-1077. Resnick D, Niwayama C. disk herniations: cartilaginous

Intravertebral (Schmorl’s) Fogel-

nodes. 8.
Blake

Radiology
GM, Jagathesan Dual x-ray

1978; 126:57-65.
T, Herd RJM, anteroposterior/lateral
1994; H. disease. 67:624-630. The human 2nd ed. New

man I.
9. studies. Schmorl spine

BrJ Radiol G, Junghanns in health and

York, 10.
Resnick

NY:

Grune

& Stratton,

1971,

3-41.

D, Niwayama

C.

Osteoporosis.

In: Resnick
of bone delphia,

D, Niwayama

C, eds. Diagnosis

and joint disorders. 2nd ed. PhilaPa: Saunders, 1988; 2023-2065.

256

#{149} Radiology

January

1997


				
DOCUMENT INFO
Shared By:
Stats:
views:823
posted:1/21/2010
language:Malay
pages:4
Description: Cupids Bow Contour ofthe Vertebral Body Evaluation