Journal of Pediatric Orthopaedics
20:709–717 © 2000 Lippincott Williams & Wilkins, Inc., Philadelphia
Diagnosing Childhood Acetabular Dysplasia Using the Lateral
Margin of the Sourcil
Hui Taek Kim, M.D., Jung Il Kim, M.D., and Chong Il Yoo, M.D.
Study conducted at Pusan National University Hospital, Pusan, Korea
Summary: The purpose of this study was to evaluate the lat-
eral edge of the acetabulum and locate the most accurate mark-
ing point on a plain radiograph when measuring the acetabular
index (AI) and the center-edge (CE) angle. We studied the
radiographs of 53 patients with unilateral developmental dys-
plasia of the hip (DDH), all of whom were treated by closed
reduction. In addition to plain radiographs, eight patients had
magnetic resonance imaging (MRI) studies, 16 had three-
dimensional computed tomography (3-DCT) studies, and six
The availability of diagnostic imaging modalities,
such as ultrasonography, magnetic resonance imaging
(MRI), and three-dimensional computed tomography (3-
DCT), has enabled physicians to more accurately diag-
nose and better treat acetabular dysplasia (9,11–16).
Each modality has its advantages and disadvantages.
Ultrasonography is helpful in the diagnosis and treat-
ment of developmental dysplasia of the hip (DDH) in
infants younger than 3 months of age (5). MRI is ex-
tremely useful in studying cartilaginous structures of the
hip, and 3-DCT provides an accurate 3-D relationship
between structures in complex hip deformities. However,
the high cost of MRI and 3-DCT studies, along with the
increased radiation exposure in 3-DCT, limits their use in
long-term follow-up, which is required in the manage-
ment of childhood acetabular dysplasia (7,10). Thus,
plain radiographs still maintain an important role in the
management of this patient population (2,3).
In the treatment of DDH, the acetabular index (AI) and
the center-edge (CE) angle have traditionally been used
as diagnostic factors for hip dysplasia. In a child one can
usually visualize both the anterior and posterior borders
of the acetabular margin, which overlap. The lateral scle-
rotic tip is the point that should be selected for measuring
Address correspondence and reprint requests to Dr. H. T. Kim, De-
partment of Orthopaedic Surgery, Pusan National University Hospital,
1Ga-10, Ami-Dong, Seo-Gu, Pusan 602-739, Korea. E-mail:
kimht@hyowon.cc.pusan.ac.kr
From the Department of Orthopaedic Surgery, Pusan National Uni-
versity Hospital, Pusan, Korea.
709
underwent arthrographic studies. We found that the most lateral
bony margin of the acetabular roof on plain radiograph repre-
sents the anterolateral portion of the acetabulum. The lateral
end of the sourcil indicates the lateral margin of the mid-
superior portion of the acetabulum. To reduce intra- and in-
terobserver errors, we suggest that when measuring the AI and
the CE angle, physicians clearly indicate in the medical records
which of the two marking points was used. Key Words: De-
velopmental dysplasia of the hip (DDH)—Sourcil.
the AI and CE angle. However, it is sometimes difficult
on plain radiographs to distinguish precisely the lateral
edge of the acetabulum to measure accurately the AI and
the CE angle because of indistinct bony shadows, espe-
cially in acetabular dysplasia. Under these circum-
stances, measuring the AI and the CE angle may result in
intra- and interobserver errors and produce erroneous
values.
We have studied the lateral edge of the acetabulum in
pediatric dysplastic hips using plain radiography, MRI,
3-DCT, and arthrography. On plain radiographs, we have
frequently noted that the shape of the sourcil varies, dif-
fering with the patient’s age and the position in which the
radiograph was taken. In a normal young child’s hip,
double radiographic shadows can overlap the sourcil
(Fig. 1A) and the sourcil can also be seen as a single,
dense straight line (Fig. 1B). The normal sourcil usually
has a smooth configuration and its orientation is hori-
zontal. However, in the dysplastic hip, the sourcil has an
irregular, short, interrupted shape with an upward slope.
The lateral end of the sourcil is incompletely rendered
and does not extend to the lateral bony margin of the
acetabular roof (Fig. 1C,D).
The purpose of this study was to clarify the anatomical
relationship between the sourcil and overlapping bony
shadows on plain radiographs and to evaluate the lateral
edge of the acetabulum to locate the most accurate mark-
ing point. In addition, we assessed the difference in the
measurement of the AI and the CE angle using two dif-
ferent measuring points: the most lateral bony margin of
the acetabulum and the lateral end of the sourcil, noting
710 H. T. KIM ET AL.

FIG. 1. A: Radiograph of a normal hip in a 2-year-old girl. Double radiographic shadows are seen in the acetabulum. The sourcil is seen
as a curved area of dense bone (arrow). Another radiographic shadow overlaps the sourcil. It is well known that these shadows may vary
depending on the rotation of the pelvis. B: Radiograph of a normal hip in an 8-month-old boy. The sourcil is seen as a single, dense,
straight line (arrow). Anterior and posterior borders may overlap and make measurement of AI and CE angle much more straightforward.
C: Radiograph of a dysplastic hip in a 6-year-old boy. The sourcil is not clearly delineated and is shorter than normal (arrow). The lateral
end of the sourcil is incompletely rendered and does not extend to the lateral bony margin of the acetabular roof. D: Radiograph of a
dysplastic hip in a 2-year-old girl 1 year after closed reduction of DDH. The sourcil is irregular in shape (arrow). A radiographic shadow
overlaps the sourcil. The slope of the sourcil is greater than that found in a normal hip.

in particular the feasibility of using the lateral end of the
sourcil as a measuring point.
MATERIALS AND METHODS
We reviewed the plain radiographs of 53 patients who
had been treated for DDH. Requirements for inclusion in
the study were unilateral DDH that was treated by closed
reduction, so that we could compare the AI and CE angle
with the uninvolved hip. In this study, we focused on the
analysis of measurement in dysplastic hip. The AI and
CE angle were measured on the radiographs by two
methods: method A, using the most lateral bony margin
of the acetabulum and method B, using the lateral end of
the sourcil (Fig. 2A,B). The AI was measured when the
patients were 1, 2, 4, 6, and 8 years of age, and the CE
angle was measured when the patients were 4, 6, 8, 10,
and 12 years of age.
In addition to the radiographic studies, four patients
underwent MRI studies and 16 patients had 3-DCT stud-

J Pediatr Orthop, Vol. 20, No. 6, 2000

 DIAGNOSIS OF ACETABULAR DYSPLASIA 711

FIG. 2. Two methods of measuring the AI and the CE angle. A: Method A involves using the most lateral bony margin of the acetabulum.
The AI is measured as the angle formed by lines A and H, and the CE angle is measured as the angle formed by line PCA. B: Method
B involves using the lateral end of the sourcil. The AI is measured as the angle formed by lines B and H, and the CE angle is measured
as the angle formed by line PCB. H, Hilgenreiner’s line; C, center of femoral head.

ies to obtain a greater anatomical understanding of the RESULTS

sourcil and the overlapping bony shadow seen on plain
radiograph. The MRI scans were obtained using a
Siemens Magneton 1.5-T Superconductive system (Mag-
neton Vision; Siemens, Erlangen, Germany). The
3-DCT studies were performed using a General Electric
scanner (LightSpeed QX/i; General Electric, Milwaukee,
WI) and a CEMAX 5000 imaging processor. Although
the MRI and 3-DCT studies were not performed in the
same patients, the data in each of the studies were con-
sidered sufficient to provide useful information regard-
ing hip morphology, especially that of the sourcil and the
lateral edge of the acetabulum.
Six patients also underwent arthrographic studies,
which were performed before consideration of surgery or
at the time of surgical treatment. In the arthrographic
studies, the location of the lateral margin of the acetabu-
lum was confirmed by insertion of a spinal needle from
a lateral approach to the hip joint during arthrography.
The sourcil was also evaluated intraoperatively during
pelvic osteotomy.
Wilcoxon signed-rank tests were used to analyze the
data with a p <0.05 considered statistically significant.
Measurement of the AI and CE angle on plain
radiographs using two methods of calculation
Table 1 summarizes the differences in the measured
AI using methods A (lateral bony margin of the acetabu-
lum) and B (lateral end of the sourcil). At 1 year of age,
the mean AI was 31.9° using method A and 39.0° using
method B, a statistically significant difference of 7.1° (p
< 0.05). The absolute values of the AI and the difference
in measured values using the two measuring methods
gradually decreased as the children matured owing to
acetabular cartilage ossification; however, at 8 years of
age, the difference between the two methods of 2.9° was
still statistically significant (p < 0.05).
Table 2 summarizes the differences in the measured
CE angle using methods A and B. At 1 year of age, the
mean CE angle was 13.8° using method A and 8.0° using
method B, a statistically significant difference of 5.8° (p
< 0.05). The absolute values of the CE angle gradually
increased and the difference in measured values of the
CE angle using the two measuring methods gradually
decreased as the children matured; however, at 12 years

TABLE 1. Measurement of the AI (mean ± SD) in TABLE 2. Measurement of the CE angle (mean ± SD) in
dysplastic hips using two methods of calculation dysplastic hips using two methods of calculation
AI (deg) Difference CE angle (deg) Difference
Age (y)/no. Method A Method B Degrees Significance (p) Age (y)/no. Method A Method B Degrees Significance (p)
1/(n ⳱ 53) 31.9 ± 4.6 39.0 ± 6.7 7.1 ± 4.3 <0.001 4/(n ⳱ 53) 13.8 ± 7.3 8.0 ± 6.7 5.8 ± 7.4 <0.001
2/(n ⳱ 50) 28.0 ± 4.3 33.4 ± 6.3 5.4 ± 4.5 <0.001 6/(n ⳱ 48) 16.4 ± 7.2 12.3 ± 6.7 4.1 ± 4.4 <0.001
4/(n ⳱ 53) 25.0 ± 4.3 30.1 ± 6.2 5.1 ± 4.6 <0.001 8/(n ⳱ 45) 18.2 ± 5.7 14.2 ± 5.4 4.0 ± 1.8 <0.001
6/(n ⳱ 48) 24.2 ± 4.2 27.3 ± 6.7 3.1 ± 3.3 <0.001 10/(n ⳱ 53) 19.8 ± 4.3 17.3 ± 5.1 3.5 ± 2.2 <0.001
8/(n ⳱ 45) 23.2 ± 4.7 26.1 ± 7.7 2.9 ± 2.7 <0.001 12/(n ⳱ 50) 20.5 ± 4.5 18.2 ± 4.1 2.3 ± 1.4 <0.001

SD, standard deviation; n, number of hips. SD, standard deviation; n, number of hips.

J Pediatr Orthop, Vol. 20, No. 6, 2000

712 H. T. KIM ET AL.

FIG. 3. A: Radiograph of a 4-year-old girl 3 years after treatment

of left DDH by closed reduction. The sourcil is irregular in shape
and its orientation is oblique (arrow). A radiographic shadow
overlaps the sourcil. The AI is 20° when measured using the most
lateral bony margin of the acetabulum (method A). However,
when the lateral margin of the sourcil is used (method B), the AI
is 38°. B: MRI coronal section of the anterior one third of the left
hip joint (arrow) indicates a relatively well-developed acetabular
roof shadow compared with that of the normal right hip. C: MRI
coronal section of the middle one third of the same left hip joint
indicates a large mid-superior bony defect (arrow). The shape of
the acetabular roof in this image coincides with that of the sourcil
on the plain radiograph (A). D: MRI sagittal section (T1WI) of the
lateral margin of the acetabulum shows a mid-superior defect
(arrow) of the left (L) acetabulum compared with that of the nor-
mal right (R) hip. E: Radiograph performed 2 years after Pem-
berton acetabuloplasty and femoral varus osteotomy.

J Pediatr Orthop, Vol. 20, No. 6, 2000

 DIAGNOSIS OF ACETABULAR DYSPLASIA 713

that a thick lateral cartilaginous acetabular margin was

FIG. 3. Continued
of age, the difference of 2.3° was still statistically sig-
nificant (p < 0.05).
MRI study
In eight hips, MRI revealed the relationship between
the cartilaginous acetabulum and the bony acetabulum.
The most valuable information of MRI in this study was
attenuated and does not appear to have developmental
potential. The shape of an as-yet unossified acetabular
edge closely mirrors the shape of the bony acetabular
rim. In addition, MRI studies provided evidence that a
short, irregular sourcil with an oblique orientation on
plain radiograph indicated a mid-acetabular bony defect
(Fig. 3). Coronal and sagittal sections of the middle one
third of dysplastic hip joints demonstrated a mid-superior
acetabular defect that was not easily recognized on plain
radiographs. The shape of the acetabular roof on coronal
sections of the middle one third of dysplastic hips coin-
cided with that of the shape of the sourcil as seen on plain
radiographs.
3-DCT studies
3-DCT studies allowed us to identify various types of
acetabular deficiency (minimal deficiency, anterosupe-
rior deficiency, mid-superior deficiency) in DDH. Mid-
superior deficiency could be distinguished only on the
lateral 3-DCT view because the AP radiograph was very
similar in all dysplastic hips.
True lateral, 90° rotated 3-DCT images of the acetabu-
lum, with the femoral head subtracted, provided an ac-

FIG. 4. A: Radiograph of a 6-year-old girl with an untreated dislocation of the right hip. The sourcil (arrowheads) is irregular in shape and
its orientation is upward. The AI is greater than that found in a normal hip. B: Coronal reformation 2-DCT image of the mid-portion of the
acetabulum shows that the shape of the bony margin of the acetabular roof is oblique, which coincides with the orientation of the sourcil
on plain radiograph (A) and that the AI is increased. C: Lateral 3-DCT image of the right hip, with the femoral head subtracted, better
demonstrates the inner side of the acetabulum. In this image, we can see a large, mid-superior defect of the acetabulum, which
corresponds to a false acetabulum. The bony margin of the inner side of the mid-acetabulum (arrowheads) corresponds to the sourcil
shape seen on the plain radiograph (A). An anterolateral bony shadow (arrow) corresponds to the bony shadow overlapping the sourcil
on the plain radiograph (A). D: Radiograph performed 1 year 6 months after open reduction, capsulorrhaphy, femoral varus derotational
osteotomy with femoral shortening, and Salter innominate osteotomy of the right hip.

J Pediatr Orthop, Vol. 20, No. 6, 2000

714 H. T. KIM ET AL.

curate 3-D morphology of the acetabulum. For example,
in untreated DDH, the bony margin of the inner side of
the mid-acetabulum on the lateral 3-DCT image (arrow-
heads in Fig. 4C) corresponded to the shape of the sourcil
on plain radiograph (arrowheads in Fig. 4A). The antero-
lateral bony shadow of the acetabulum on 3-DCT (arrow
in Fig. 4C) corresponded to the bony shadow overlap-
ping the sourcil on plain radiograph (arrow in Fig. 4A).
Because this portion of the acetabulum is near the an-
teroinferior iliac spine, measurement of the AI and the
CE angle using this anterolateral bony shadow on plain
radiograph could not reflect a mid-superior acetabular
deficiency.
Arthrographic study
Absolute values of this arthrographic study have yet to
be reinforced by other sophisticated diagnostic tools,
which make it possible to calculate 3-D coordinates of
points on the edge of the acetabulum.
During arthrography of the hip, a spinal needle was
inserted into the anterior and into the middle portions of
the acetabulum from a lateral approach to the hip joint
(Figs. 5 and 6). The tip of the needle that was inserted to
the anterior portion of the acetabulum was placed at the
most lateral bony shadow of the acetabulum, which cor-
responds to the anterolateral portion of the acetabulum
(Fig. 5B). The tip of the needle that was inserted to the
middle portion of the acetabulum was placed at the lat-
eral end of the sourcil, which corresponds to the mid-
superior portion of the acetabulum (Fig. 5C and Fig. 6C).
These findings confirmed that the lateral end of the
sourcil reflects a mid-acetabular portion on plain radio-
graphs.
Intraoperative study
During the Salter innominate osteotomy in two dys-
plastic hips, the indicator placed in the mid-superior por-
tion of the acetabulum indicated the lateral margin of the
sourcil (Fig. 7).
DISCUSSION
The sourcil is a curved area of dense bone on the
weight-bearing surface of the acetabulum. This radio-
graphic shadow indirectly represents a stress distribution
within the hip joint (1). In a normal hip, the sourcil is
uniformly thick and semilunar in shape, with a horizontal
or downward orientation. However, in the dysplastic hip,
the orientation of the sourcil is upward, suggesting an
uneven distribution of stress within the hip joint.

FIG. 5. Plain radiograph and arthrographic images of a 7-year-old boy. The patient had a mild, waddling gait and difficulty in climbing
stairs. His mother had been diagnosed with dysplastic hips and treated with osteotomy. A: Radiograph shows an obscure sourcil, which
is shorter than normal. The lateral end of the sourcil does not extend to the lateral bony margin of the acetabular roof. B: In this
intraoperative arthrographic image, a spinal needle is inserted into the lateral aspect of the hip joint to the anterior portion of the
acetabulum. The tip of the needle is placed at the most lateral bony shadow of the acetabulum, which corresponds to the anterolateral
portion of the acetabulum. C: In this intraoperative arthrographic image, a spinal needle is inserted into the lateral aspect of the hip joint
to the middle portion of the acetabulum. The tip of the needle is placed at the lateral end of the sourcil, which corresponds to the
mid-superior portion of the acetabulum.

J Pediatr Orthop, Vol. 20, No. 6, 2000

 DIAGNOSIS OF ACETABULAR DYSPLASIA 715

FIG. 6. Plain radiographs and arthrographic image of a 5-year-old girl with right DDH. A: Anteroposterior radiograph in neutral position
shows the sourcil is irregular in shape and its orientation is upward in slope (arrow). B: Radiograph with hip abducted and internally rotated
(the image is the same as the caudal view of the pelvis). The sourcil of the right hip is short and interrupted (arrow) compared with that
of the normal left hip, and its orientation is upward in slope. C: Intraoperative arthrographic image shows the spinal needle inserted in the
lateral edge of the sourcil, which is the mid-portion of the acetabulum. D: Radiograph performed 2 years after Salter innominate osteotomy
to correct dysplasia.

The shape of the sourcil in childhood hip dysplasia is
difficult to discern accurately because large portions of
the acetabulum are cartilaginous. For this reason, numer-
ous other methods have been used to measure childhood
acetabular dysplasia, including AI (21), CE angle (22),
femoral head uncovering (6), head–teardrop distance
(19), Smith’s c/b and h/b ratios (20), ACM angle and
M-Z distance (21), and center-head distance discrepancy
(CHDD) (4).
Among these methods, two popular parameters that
use the acetabular edge as a measurement point are the
AI and the CE angle. The AI is measured between the
horizontal Hilgenreiner’s line and a line extending from
the superolateral margin of the triradiate cartilage to the
most lateral ossified margin of the acetabulum. However,
this method is only viable in measuring development of
the hip in children who are younger than 8 years of age,
because Hilgenreiner’s line is difficult to measure after
ossification of the triradiate cartilage. The CE angle is a
useful method only in children after 5 years of age, be-
cause the center of the femoral head is difficult to define
in a younger child because of an eccentrically located
ossific nucleus.
In addition, the traditional measurement of a 3-D
structure on standardized 2-D plain radiographs does not
appear to reflect adequately the variability and complex-
ity of the deformity, nor does it permit us to measure
accurately the deformity in a reproducible manner. When
measuring the AI, errors can occur owing to the incorrect
positioning of the child for radiographs. Errors of ±3°
can be caused either by pelvic flexion/extension or by
pelvic rotation. With increasing lumbar lordosis, which
corresponds to the extent of pelvic extension, the AI
increases, and with decreasing lordosis or pelvic flexion,
the AI decreases. Likewise, if the pelvis is rotated about
the longitudinal body axis, the AI toward the side of the
rotation decreases and the AI on the opposite side in-
creases (18,21).
Furthermore, during measurement of the AI and the
CE angle, it is difficult to mark the exact lateral bony
margin of the acetabulum because of irregular and indis-
tinct bony shadows on radiographs of dysplastic hips
(8,22). In the dysplastic hip, the lateral end of the sourcil
is incompletely rendered, as well. Another variable bony
shadow, which extends more laterally over the lateral
end of the sourcil, often overlaps the sourcil. In these
circumstances, the AI and the CE angle are poorly re-
producible.
In this study, we attempted to clarify the anatomical
relationship between the sourcil and overlapping bony

J Pediatr Orthop, Vol. 20, No. 6, 2000

716 H. T. KIM ET AL.

FIG. 7. A: Anteroposterior radiograph of a 7-year-old girl with a left DDH shows a poorly developed, short, oblique sourcil (arrow). She
had been treated by closed reduction at the age of 13 months. B: Intraoperative arthrographic image of same hip. C: After Salter
innominate osteotomy, the orientation of the sourcil is more horizontal in slope. Indicator (asterisk) placed in the mid-superior portion of
the acetabulum during the operation demonstrates the lateral margin of the sourcil. D: Radiograph performed 3 years after Salter
innominate osteotomy.

shadows on plain radiographs. We found that the radio-
graphic lateral bony margin of the acetabulum extends
beyond the lateral end of the sourcil on plain radiographs
and that this shadow corresponds to the anterolateral
margin of the acetabulum. The lateral end of the sourcil,
which is located more medial to this shadow, corre-
sponds to the lateral edge of the mid-superior portion of
the acetabulum.
In children, there is an ossification center on the lateral
portion of the acetabulum that is associated with acetab-
ular growth. This nucleus begins ossification when the
child is about 9 years of age and fuses to the ilium when
the child is about 15 years of age (15,16). During the
growth period, continuous compression force on the os-
sifying nucleus will result in a growth disturbance of the
lateral edge of the acetabulum (17). These findings are
evident in the mid-superior acetabular dysplasia, in
which the increased magnitude of the shearing force pre-
vents normal acetabular development.
Even though the primary purpose of this study was not
to assess intra- and interobserver errors in measuring the
AI and the CE angle, we found that there is a significant
difference in measurement values on plain radiographs in
young children depending on whether one uses the most
lateral bony margin of the acetabulum or the lateral end
of the sourcil as a marking point.
In conclusion, when diagnosing childhood acetabular
dysplasia on plain radiographs, especially with measure-
ments using the lateral edge of the acetabulum (e.g., the
AI or CE angle), physicians must consider that the most
lateral bony shadow, which extends over the lateral end
of the sourcil, represents the anterolateral portion of the
acetabulum. Measurement using the lateral end of
the sourcil as a marking point represents the status of the
midsuperior portion of the acetabulum. To reduce intra-
and interobserver errors, we suggest that when measur-
ing the AI and the CE angle, physicians clearly indicate
in themedical records which of the two marking points was
used.
Acknowledgment: We thank Dennis R. Wenger, M.D.,
from the Children’s Hospital and Health Center–San Diego,
University of California, and Daniel R. Faber for their kind
help with this article.

J Pediatr Orthop, Vol. 20, No. 6, 2000

 DIAGNOSIS OF ACETABULAR DYSPLASIA
REFERENCES
1. Bombelli R. Structure and function in normal and abnormal hips:
how to rescue mechanically jeopardized hips. 3rd ed. New York:
Springer, 1993:123–44.
2. Broughton NS, Brougham DI, Cole WG. Reliability of radiological
measurement in the assessment of the child’s hip. J Bone Joint
Surg Br 1989;71:6–8.
3. Brougham DI, Broughton NS, Menelaus MB. The predictability of
acetabular development after closed reduction for congenital dis-
location of the hip. J Bone Joint Surg Br 1988;70:733–6.
4. Chen IH, Kuo KN, Lubicky JP. Prognosticating factors in acetab-
ular development following reduction of developmental dysplasia
of the hip. J Pediatr Orthop 1994;14:3–8.
5. Clarke NMP, Harke HT, McHugh P, et al. Real-time ultrasound in
the diagnosis of congenital dislocation and dysplasia of the hip. J
Bone Joint Surg Br 1985;67:406–12.
6. Dickens DRV, Menelaus MB. The assessment of prognosis in
Perthes’ disease. J Bone Joint Surg Br 1978;60:189–94.
7. Fisher R, O’Brien TS, Davis KM. Magnetic resonance imaging in
congenital dysplasia of the hip. J Pediatr Orthop 1991;11:617–22.
8. Harris NH. Acetabular development in congenital dislocation of
the hip. J Bone Joint Surg Br 1975;57:46–52.
9. Harris NH. Acetabular growth potential in congenital dislocation
of the hip and some factors upon which it may depend. Clin Orthop
1976;119:99–106.
10. Kim HT, Wenger DR. The morphology of residual acetabular de-
ficiency in childhood hip dysplasia: three-dimensional computed
tomographic analysis. J Pediatr Orthop 1997;16:637–47.
717
11. Lindstrom JR, Ponseti IV, Wenger DR. Acetabular development
after reduction in congenital dislocation of the hip. J Bone Joint
Surg Am 1979;61:112–7.
12. Malvitz TA, Weinstein SL. Closed reduction of congenital dyspla-
sia of the hip. J Bone Joint Surg Am 1994;76:1777–92.
13. Murphy SB, Kijewski PK, Millis MB, et al. Acetabular dysplasia
in the adolescent and young adults. Clin Orthop 1990;201:214–23.
14. Murphy SB, Ganz R, Müller ME. The prognosis in untreated dys-
plasia of the hip. J Bone Joint Surg Am 1995;77:985–9.
15. Ponseti IV. Growth and development of the acetabulum in the
normal children. J Bone Joint Surg Am 1978;60:575–85.
16. Ponseti IV. Morphology of the acetabulum in congenital disloca-
tion of the hip. J Bone Joint Surg Am 1978;60:586–99.
17. Portinaro NMA, Matthews SJE, Benson MKD. The acetabular
notch in hip dysplasia. J Bone Joint Surg Br 1994;76:271–3.
18. Portinaro NMA, Murray DW, Bhullar TPS, et al. Errors in mea-
surement of acetabular index. J Pediatr Orthop 1995;15:780–4.
19. Scoles PV, Boyd A, Jones PK. Roentgenographic parameters of
the normal infant hip. J Pediatr Orthop 1987;7:656–63.
20. Smith WS, Badgley CE, Orwig JB, et al. Correlation of postre-
duction roentgenograms and thirty-one-year follow-up in congen-
ital dislocation of the hip. J Bone Joint Surg Am 1968;50:1081–98.
21. Tönnis D. Normal value of the hip joint for the evaluation of
X-rays in children and adults. Clin Orthop 1976;119:39–47.
22. Wiberg G. Studies on dysplastic acetabula and congenital subluxa-
tion of the hip joint: with special reference to the complications of
osteoarthritis. Acta Chir Scand 1939;83(suppl 58).
J Pediatr Orthop, Vol. 20, No. 6, 2000