418 Original article
Natural progression of hip dysplasia in newborns: a reflection
of hip ultrasonographic screenings in newborn nurseries
Hui-Wen Chena,d, Chia-Hsieh Change, Suei-Tsau Tsaia, Wen-Jiunn Liub,
Ceferino Chuab, Yu-Ying Chenc and Ken N. Kuof
Hip screening is recommended for early detection
of hip dysplasia; however, the strategy and efficacy were
questioned. We performed ultrasonography to objectively
observe the hip development in a group of infants
and raise reflections on the efficacy of newborn
ultrasonographic hip screening. A total of 1333 newborns
(705 male and 628 female) received hip ultrasonography
in the first week after birth. Clinical hip instability was
c 2010 Wolters Kluwer Health | Lippincott
detected by the Ortolani test. Hip dysplasia was defined
by ultrasonography. A second survey of 90 babies was
performed 1 month later on the babies with clinical hip
instability or ultrasonographic Graf’s type IIa, IIc, D, III,
and IV hips. All the hip ultrasonographies were performed
by the same investigator. Clinical instability was detected
in 13 hips of 10 babies. Manual tests helped to detect
unstable hips that had worse ultrasonographic
measurements than those stable hips. Five Graf’s type IIc
hips and four type III hips were detected from the 2666
hips. At the age of 1 month, eight of the nine Graf’s type IIc
or III hips attained physiological status without treatment.
A strong trend of spontaneous resolution in clinical hip
Introduction
Developmental dysplasia of the hip (DDH) is the most
common musculoskeletal disorder in infants. DDH is a
correctable deformation and early splinting restores the
normal hip anatomy in more than 90% of the patients [1–3].
For early detection and treatment, hip screening by
manual tests has been the standard in newborn nurseries
and baby clinics. Despite excellent results from newborn
hip screening, as shown in some studies [4–7], incidences
of late diagnosed DDH continue to occur [8–10].
Since the first application of ultrasound to visualize infant
hips in the early 1980s [11,12], it has became an estab-
lished method to detect hip dysplasia and monitor the
course of treatment [13,14]. The use of ultrasound in
hip screening was reported to decrease the rate of late
diagnosed DDH [15–17]. However, the strategy of ultra-
sound screening, universal or selective, is still debatable
[18–21]. The optimal timing of ultrasound examination
in newborn nurseries or before 3 months of age is also
controversial [22,23].
On account of the local folk custom in Taiwan wherein
the mother and baby were expected to stay home for
1 month after birth, most of the babies with clinical
instability of the hip or abnormal hip ultrasound, which is
1060-152X
c 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins
instability and ultrasonographic dysplasia was observed
in the newborns. The results did not support immediate
treatment on the basis of newborn manual or ultrasound
hip screening. We recommend manual testing in the
newborn nursery to detect the hips at risk of dysplasia and
ultrasonography after the first month after birth to confirm the
diagnosis and judge the management. J Pediatr Orthop B
19:418–423
Williams & Wilkins.
Journal of Pediatric Orthopaedics B 2010, 19:418–423
Keywords: developmental dysplasia of the hip, neonatal screening,
ultrasonography
Departments of aPediatrics, bOrthopedics, cChild Rehabilitation Center, Taiwan
Adventist Hospital, dDepartment of Pediatrics, Buddhist Tzu Chi General
Hospital, Taipei, eDepartment of Pediatric Orthopedics, Chang Gung Memorial
Hospital, Chang Gung University, Taoyuan and fCenter for Health Policy Research
and Development, National Health Research Institute, Miaoli, Taiwan
Correspondence to Ken N. Kuo, MD, Center for Health Policy R&D, National
Health Research Institute, 35 Keyan Road, Zhunan Town, Miaoli County 35053,
Taiwan
Tel: + 886 37 246166 x36300; fax: + 886 37 586261;
e-mail: kennank@aol.com
detected in a newborn nursery, had no treatment in the
first month after birth. This offered an opportunity to
observe the natural outcome of neonatal hip instability
and ultrasonographic dysplasia. The aim of this observa-
tional study is to offer evidence from a group of accidentally
untreated patients and raise reflections on the efficacy of
newborn hip screening.
Patients and methods
From April 2006 to December 2007, all women who gave
birth in the authors’ hospital were informed of hip
ultrasonographic study. Procedures and possible problems
were explained by one of the authors. Participation in the
study was voluntary. The medical records of babies who
received hip ultrasonography in the first week after birth
were reviewed. The study was approved by the hospital’s
institutional review board.
The newborns were placed in a supine position without
sedation. The stability of the hip joints was evaluated
manually by the Ortolani test [24]. For the ultrasono-
graphic study, the hips of the newborns were examined
in slight flexion and neutral in axial rotation and
abduction-adduction. A linear-array ultrasound transducer
was applied horizontally at the lateral aspect of the hip.
DOI: 10.1097/BPB.0b013e328339ecff
 Natural progression of hip dysplasia in newborns Chen et al. 419

The proximal femur and straight line of the ilium were
identified to determine the plane crossing the mid-
portion of the acetabulum.
Graf ’s a and b angles [11,12] were measured. The
a angle, an indicator of acetabular development by
measuring the bony roof, is defined as the angle between
the iliac line and the line from the bony edge of the
acetabulum superior laterally to the lowest point of the
ilium inferior medially. The b angle, an indicator of hip
displacement by measuring the cartilaginous roof, is
defined as the angle between the iliac line and the line
from the bony edge of the acetabulum to the labrum
laterally.
All hip ultrasonographies were performed by one pedia-
trician who was trained to perform musculoskeletal ultra-
sonographic evaluation. We randomly chose 20 successive
newborns to test the reliability of ultrasonographic
measurement. The measurement of the a and b angles
had a good test–retest reliability (intraclass correlation
coefficient, 0.785 and 0.800, respectively).
Babies with clinical hip instability or ultrasonographic
dysplasia (Graf type IIc, D, III, or IV) were referred to
orthopedic surgeons for further management after being
discharged from the nursery. A second hip ultrasonogra-
phy was performed 4–6 weeks after birth for the above
babies and babies with Graf ’s IIa hips by the same
investigator. Babies with persistent hip pathology on the
second ultrasonography were referred to orthopedic
surgeons again. Pelvis radiography was performed at the
age of 4 months to confirm the hip condition.
The a and b angles were compared between the new-
borns grouped by sex, birth body weight, gestational age,
birth order, and breech. The ultrasonographic findings in
the hips with a positive Ortolani test were compared with
that in the hips with a negative Ortolani test. Indepen-
dent t-test was used for continuous variables. Categorical
variables (Graf ’s type) were put into a table and analyzed
with Fisher’s exact test because of the small numbers in
each cell.
Ultrasonography was performed on both hips of the 1333
babies. The mean a angle in the 2666 hips was 64.71
(range 38–781, standard deviation 3.7), and the mean
b angle was 45.51 (range 23–871, standard deviation 6.8).
According to Graf ’s classification, three babies had four
type III hips (bilateral in one baby and unilateral in two
babies), four babies had five type IIc hips (bilateral in one
baby and unilateral in three babies), and the other 82
babies had 122 type IIa hips (bilateral in 40 babies and
unilateral type IIa plus type I in 42 babies). A total of
seven babies had nine hips with ultrasonographic dys-
plasia (five type IIc and four type III).
The mean values of the a and b angles in newborns
grouped by sex, birth body weight, gestational age, birth
order, and breech presentation are listed in Table 1. Sex is
a significant factor that influences both the acetabular
development and the position of the femoral head. Low
birth body weight (< 2500 g) and prematurity (< 37 weeks)
were associated with a low b angle (Table 1).
Ten babies had a positive Ortolani test in 13 hips. The 13
unstable hips were three type III, three type IIc, six type
IIa, and one type I. The data of the 10 babies and the
results one month after ultrasonography are listed in
Table 2. On an average, the mean a angle was 53.71 and
the mean b angle was 64.81. The data were significantly
different from the mean a angle of 66.21 and the mean
b angle of 44.01 in the other 1323 babies without hip
instability (P < 0.001). Ultrasonographic dysplasia (type
IIc and beyond) was detected in six of the 20 hips (30%)
of the 10 babies with a positive Ortolani test and in three
of the 2646 hips (0.1%) in babies with a negative Ortolani
test. The rate to detect a dysplastic hip by ultrasound was
significantly greater in the babies with a positive Ortolani
test (P < 0.001). (Fig. 1)
Table 1 Mean data of a and b angles according to characters
of newborns
a angle b angle
Number of newborns (%) Mean P value Mean P value

Sex
Results Male 705 (53) 65.5 < 0.001 44.1 < 0.001
Between April 2006 and December 2007, 1333 consecu- Female 628 (47) 63.8 47.2
Birth body weight
tive newborns (705 male and 628 female) received hip Z 2500 g 1240 (93) 64.7 0.672 45.8 < 0.001a
ultrasonography in the first week after birth. The mean < 2500 g 93 (7) 64.8 42.6
gestational age was 38.3 weeks (range 33–42 weeks), and Gestational age
Z 37 weeks 1230 (92) 64.7 0.660 45.8 < 0.001b
103 babies were premature (gestational age < 37 weeks). < 37 weeks 103 (8) 64.6 43.5
Birth body weight was 3090 g on an average (range 1700– Birth order
First 830 (63) 64.7 0.581 45.4 0.342
4320 g) and 93 babies were born with a weight less than Others 503 (37) 64.7 45.7
2500 g. Of the 1333 babies, 105 (7.9%) had breech Breech
presentation and 830 (62.3%) were first births. Hip inst- Positive 105 (8) 64.2 0.055 45.3 0.545
Negative 1228 (92) 64.7 45.6
ability, tested by the Ortolani and Barlow tests, was found
a
in 10 babies (bilateral hip instability in three babies and Correlation between birth body weight and b angle was significant (Pearson’s
r = 0.102, P < 0.001).
unilateral hip instability in seven babies). No irreducible b
Correlation between gestational age and b angle was significant (Pearson’s
dislocation of the hip was detected. r = 0.137, P < 0.001).
420 Journal of Pediatric Orthopaedics B 2010, Vol 19 No 5

Management after newborn hip ultrasonography The second survey

Thirteen babies with clinical hip instability or Graf ’s type We arranged for 90 babies to return for a second survey,
IIc, III hips were referred to orthopedic clinics after including 10 babies with clinical instability and 80 babies
discharge from the newborn nursery. However, most of with ultrasound findings (type IIa and beyond) after the
them stayed home in the first month of after birth first month after birth. Two babies with a unilateral
because of the local folk custom. Two of the 13 babies Ortolani-positive hip underwent harness therapy by
were brought to orthopedic clinic and had harness orthopedic doctors and did not come back for the second
treatment in the first month after birth. Both of them ultrasound. Twelve babies with type IIa hips missed the
had unilateral type IIa hips. Another baby initially had second survey. There were 76 babies, eight babies for
bilateral type IIc hips and improved to one type IIa hip clinical instability and 68 babies for ultrasound findings,
and one type I hip 1 month later. Clinical instability who came back to undergo ultrasound survey. They did
was absent. This baby received harness treatment at the not have any treatment in the first month of life (Fig. 1).
age of 3 months because of unilateral hip dysplasia shown The Ortolani test was negative in all babies at age of 1
on the pelvis radiograph. Of the 1333 babies, three month. In the average 4.2 weeks between the two
underwent harness treatment; the rate of treatment ultrasonographic studies, the mean a angle of the 76
was 0.23%. babies improved from 55.6 to 64.81 and the b angle
improved from 58.4 to 41.31. There were nine hips with
Table 2 The Graf’s classification of the 20 hips in the 10 babies ultrasonographic dysplasia at birth. The four type III hips
who had at least one unstable hip. Two babies received harness at birth became two type I hips and two type IIa hips at
therapy and the other eight babies had stable hips spontaneously 1 month of age. (Fig. 2) Of the five type IIc hips, four
1 month later
improved to type I and IIa hips, whereas one remained a
At birth At 1 month old
type IIc hip. The baby with a persistent type IIc hip was
Babies Unstable hips Stable hips Unstable hips Stable hips referred to the orthopedic clinic again. No treatment was
1 III I — I/I
recommended by orthopedic doctor because the hip was
2 III/III — — IIa/I stable on manual testing. Pelvis radiography was negative
3 IIc IIa — IIc/I at follow-up 3 months later.
4 IIc IIa — I/I
5 IIc/I — — I/I In the 82 babies with 122 type IIa hips at birth, five
6 IIa/IIa — — I/I
7 IIa I — IIa/I
babies had unstable (Ortolani test positive) type IIa hips
8 IIa I — I/I and the other 77 babies had stable type IIa hips. Two of
9 IIa IIa Harness the five babies with unstable IIa hips received harness
10 IIa I Harness
treatment. The other three babies had a second survey

Fig. 1

1333 babies

(a) Ortolani (+) Ortolani (−)

10 babies 1323 babies

III 3 hips; IIc 3 hips III 1 hip; IIc 2 hips

IIa 9 hips; I 5 hips IIa 115 hips
I 2528 hips

8 babies 2 babies (2 IIa. 21) (b) 68 babies (c) 12 babies 1243 babies
(III:3; IIc:3; IIa:7; I:3) had harness in the (III:1; IIc:2; IIa:100: (15 IIa, 9 I) (bilateral type I
second survey first month I:33) Miss second survey hips)

III:0; IIc:1, IIa:2, I:13 III:0; IIc:0, IIa:5, I:131

1 baby had harness No hip problem checked

at 3 months old at 19 months old

A diagram showing the outcome of the 1333 babies. Babies with clinical instability (cell a) and babies with clinical stable but ultrasonographic type II,
III and IV hips (cells b and c) were arranged to have the second survey. Twelve babies missed the second survey (cell c).
 Natural progression of hip dysplasia in newborns Chen et al. 421

Fig. 2

A female baby with bilateral Graf’s type III hips and a positive Ortolani test. (a) Right hip ultrasonography at birth showed the a angle of 421 and the
b angle of 811. (b) Left hip a angle was 411 and the b angle was 771. (c) Ultrasonography 1 month later showed right hip a angle of 531 and b angle
of 491 without treatment. (d) Left hip a angle was 621 and b angle was 411.

and their hips became stable type I hips. In the 77 babies
with stable IIa hips, 65 babies with 100 type II hips had
a second survey. The 100 stable IIa hips improved to 97
type I hips (97%) and to three type IIa hips 1 month later.
The other 12 babies who missed the second survey were
followed at a mean age of 19 months (range 9–29
months). There was no hip problem noted.
Discussion
A strong trend of spontaneous resolution of newborn
ultrasonographic dysplasia was observed in this study.
The high rate of spontaneous resolution in this study
was not the result of overdiagnosis. The data of the
2666 ultrasonographies were continuous variables and
distributed in a bell-shaped curve. (Fig. 3) Using the
definition of a dysplastic hip, by an a angle less than 501,
we only screened nine dysplastic hips from a total of 2666
hips (0.3%). It is reasonable to pay special attention to
such cases, and harness therapy was recommended in
Graf ’s type IIc and type III hips [12]. However, all but
one of the hips recovered to physiological status 1 month
later without treatment. The study results did not
support immediate treatment on the basis of newborn
ultrasound hip screening.
422 Journal of Pediatric Orthopaedics B 2010, Vol 19 No 5
Fig. 3
400
300
Number of hips
200
100
0 is that the ultrasound hip screening in the authors’
40 50 60 70
Alpha angle
Distribution of the a angle of the 2666 hips. The vertical line indicated
491 and the cases left to the line (ultrasonographic dysplasia) were only
0.3%.
Neonatal hip screening, either by manual test or by
ultrasound might detect a dynamic physical status in the
first few days after birth. Evidences from Barlow’s original
study showed that 60% of Barlow-positive hips became
stable after the first week and 88% of these unstable hips
became stable in the first 2 months after birth without
treatment [25]. In this study, we defined neonatal hip
instability by the Ortolani test, a strict test depending
on the relocation of a dislocated hip. It is the reason
why only 10 of the 1333 newborns (0.75%) had neonatal
hip instability. Nevertheless, the eight babies who stayed
home for 1 month resolved to have stable hips spon-
taneously. In a systemic review of literature using the best
evidence approach by Shipman et al. [26], the high rate of
spontaneous resolution of neonatal hip instability and
dysplasia was emphasized and the net benefit of hip
screening was challenged.
There were several prospective studies to compare the
results of general ultrasound screening and selective
ultrasound [18–21]. The case number in this study was
too small to make any contribution to this debatable
issue. This study is to question the immediate treatment
approach when DDH is detected in the first week after
birth. As the treatment for DDH at birth is not an urgent
requirement, the reason to support newborn ultrasound
screening is questionable.
We support manual screening in a newborn nursery
because the simple manual tests helped to detect
unstable hips that had worse ultrasonographic measure-
ments than the stable hips. Similar findings that showed
that ultrasonography in hips with clinical instability had a
higher positive rate were reported before [22,27,28]. Hip
ultrasonography could be deferred after the first month
after birth to confirm the diagnosis and judge the further
management, when the physical status became settled
and harness could still treat the hip problem.
However, one should be cautious of type III and IV hips
as they may have a worse course, in spite of our small
number of cases that showed good progression in the ob-
servation follow-up. In a country that can offer universal
ultrasound hip screening, it is more specific to perform
the first ultrasonography after the first month after birth.
A limitation of this study is that we did not follow all the
Graf ’s type I hips at birth. There could be very few cases
of DDH that developed later in life and showed a stable
and mature hip at birth. Another limitation of this study
hospital is not universal. The distribution of sex, birth
order, type of birth, and Ortolani test results in this study
group are not representative of the Taiwanese population.
The hip ultrasonography in this study differed from
Graf ’s technique in which babies were put in a lateral
position during the examination. There could be a debate
as to whether the measured a and b angles in this study
were consistent with Graf ’s normal values. However, as all
the ultrasonographies including the second survey were
performed by one investigator, the trend of improvement
in ultrasonographic measurements should be consistent
and reliable.
The study results do not support immediate treatment
on the basis of newborn hip screening because of the
strong trend of spontaneous resolution of newborn hip
instability and ultrasonographic dysplasia. We recommend
manual testing in the newborn nursery to detect the hips
at risk of instability and dysplasia, and ultrasonography
after the first month after birth to confirm the diagnosis
and judge the management.
Acknowledgement
The authors have no financial relationships relevant to
this article to disclose.
References
1 Dunn PM, Evans RE, Thearle MJ, Griffiths HE, Witherow PJ. Congenital
dislocation of the hip: early and late diagnosis and management compared.
Arch Dis Child 1985; 60:407–414.
2 Filipe G, Carlioz H. Use of Pavlik harness in treating congenital dislocation
of the hip. J Pediatr Orthop 1982; 2:357–362.
3 Pavlik A. The functional method of treatment using a harness with stirrups
as the primary method of conservative therapy for infants with congenital
disloation of the hip. Clin Orthop 1992; 281:4–10.
4 Goss PW. Successful screening for neonatal hip instability in Australia.
J Pediatr Child Health 2002; 38:469–474.
5 Bernard AA, O’Hara JN, Bazin S, Humby B, Jarrett R, Dwyer NS.
An improved screening system for the early detection of congenital
dislocation of the hip. J Pediatr Orthop 1987; 7:277–282.
6 Hansson G, Nachemson A, Palmen K. Screening of children with congenital
dislocation of the hip joint on the maternity wards in Sweden. J Pediatr
Orthop 1983; 3:271–279.

7 Tredwell SJ, Bell HM. Efficacy of neonatal hip examination. J Pediatr Orthop
1981; 1:61–65.
8 Lennox IAC, McLaughlan J, Murali R. Failures of screening and management
of congenital dislocation of the hip. J Bone Joint Surg 1993; 75-B:72–75.
9 MacKenzie IG, Wilson JG. Problems encountered in the early disgnosis and
management of congenital dislocation of the hip. J Bone Joint Surg 1981;
63-B:38–42.
10 Chang CH, Chiang YT, Lee ZL, Kuo KN. Incidence of surgery in
developmental dysplasia of the hip in Taiwan. J Formosan Med Assoc 2007;
106:462–466.
11 Graf R. The diagnosis of congenital hip-joint dislocation by the ultrasonic
Combound treatment. Arch Orthop Traum Surg 1980; 97:117–133.
12 Graf R. Classification of hip joint dysplasia by means of sonography.
Arch Orthop Trauma Surg 1984; 102:248–255.
13 Morin C, Harcke HT, MacEwen GD. The infant hip: real-time US assessment
of acetabular development. Radiology 1985; 157:673–677.
14 Millis MB, Share JC. Use of ultrasonography in dysplasia of the immature
hip. Clin Orthop Relat Res 1992; 274:160–171.
15 Rosendahl K, Markestad T, Lie RT. Ultrasound screening for developmental
dysplasia of the hip in the neonate: the effect on treatment rate and
prevalence of late cases. Pediatrics 1994; 94:47–52.
16 Maxwell SL, Ruiz AL, Lappin KJ, Cosgrove AP. Clinical screening for
developmental dysplasia of the hip in North Ireland. BMJ 2002;
324:1031–1033.
17 Kries V, Ihme RN, Oberle D, Lorani A, Stark R, Altenhofen L, et al. Universal
ultrasound screening programme for developmental dysplasia of the hip in
Germany: impact on the rate of first operative procedures. Lancet 2003;
362:1883–1887.
18 Paton RW, Hossain S, Eccles K. Eight-year prospective targeted ultrasound
screening program for instability and at-risk hip joints in developmental
dysplasia of the hip. J Pediat Orthop 2002; 22:338–341.
Natural progression of hip dysplasia in newborns Chen et al. 423
19 Lewis K, Jones DA, Powell N. Ultrasound and neonatal hip screening: the
five-year results of a prospective study in high-risk babies. J Pediat Orthop
1999; 19:760–762.
20 Holen KJ, Tegnander A, Bredland T, Johansen OJ, Saether OD, Eik-Nes SH,
et al. Universal or selective screening of the neonatal hip using ultrasound?
A prospective, randomised trial of 15,529 newborn infants. J Bone Joint
Surg Br 2002; 84:886–890.
21 Rosendahl K, Markestad T, Lie RT, Sudmann E, Geitung JT.
Cost-effectiveness of alternative screening strategies for
developmental dysplasia of the hip. Arch Pediat Adol Med 1995;
149:643–648.
22 Marks DS, Clegg J, al-Chalabi AN. Routine ultrasound screening
for neonatal hip instability. Can it abolish late-presenting congenital
dislocation of the hip? J Bone Joint Surg Br 1994; 76:534–538.
23 Krismer M, Klestil T, Morscher M, Eggl H. The effect of ultrasonographic
screening on the incidence of developmental dislocation of the hip.
Int Orthop 1996; 20:80–82.
24 Ortolani M. Congenital hip dysplasia in the light of early and very early
diagnosis. Clin Orthop 1976; 119:6–10.
25 Barlow TG. Early diagnosis and treatment of congenital dislocation of the
hip. J Bone Joint Surg Br 1962; 44:292–301.
26 Shipman SA, Helfand M, Moyer VA, Yawn BP. Screening for developmental
dysplasia of the hip: a systematic literature review for the US Preventive
Services Task Force. Pediatrics 2006; 117:e557–e576.
27 Malkawi H, Asir B, Tadros F, Khasawneh Z. Sonographic image of the
newborn hip with positive Ortolani’s sign. Clin Orthop Related Res 1992;
279:138–143.
28 Paton RW, Srinivasan MS, Shah B, Hollis S. Ultrasound screening for hips
at risk in developmental dysplasia. Is it worth it? J Bone Joint Surg Br 1999;
81:255–258.