Eur J Pediatr

DOI 10.1007/s00431-017-2959-z

ORIGINAL ARTICLE

Diagnosing developmental dysplasia of the hip using the Graf

ultrasound method: risk and protective factor analysis in 11,820
universally screened newborns
Mohammad Schams 1 & Rob Labruyère 2 & Anne Zuse 3 & Mikolaj Walensi 3

Received: 13 April 2017 / Revised: 13 June 2017 / Accepted: 26 June 2017

# Springer-Verlag GmbH Germany 2017

Abstract The essential role of ultrasound examinations in
diagnosis and treatment of developmental dysplasia of the
hip (DDH) is widely accepted while the weighting and corre-
lation of protective factors and perinatal risk for DDH still
give rise to debate. Our aim was to investigate the impact of
single and twofold combined risk and protective factors on the
newborns' hip maturity, assessed with the Graf ultrasound
method. Therefore, data sets of 11,820 universally screened
newborns were analyzed. Univariate and logistic regression
analyses were performed to correlate risk and protective fac-
tors with mature or immature but appropriate for age and
pathologic hip types. Thereby, female gender (OR 4.07
(95% CI 3.01–5.51), p < 0.001), breech presentation (4.98
(3.71–6.71), p < 0.001), and positive family anamnesis (5.05
Mohammad Schams and Rob Labruyère have equal contributions and
shared first authorship.
Communicated by Mario Bianchetti
* Mohammad Schams
praxis.schams@bluewin.ch
Rob Labruyère
rob.labruyere@kispi.uzh.ch
Anne Zuse
anne.zuse@hirslanden.ch
Mikolaj Walensi
mikolaj.walensi@uk-essen.de
1
Department of Neonatology, Hirslanden Private Hospital Group,
Klinik Hirslanden, Witellikerstrasse 40, 8032 Zurich, Switzerland
2
Rehabilitation Center for Children and Adolescents, University
Children’s Hospital Zurich, Affoltern am Albis, Switzerland
3
Clinical Trial Unit, Hirslanden Private Hospital Group, Klinik
Hirslanden, Zurich, Switzerland
(3.49–7.31), p < 0.001) as well as the combination of female
gender with high birth weight (3.51 (2.45–5.03), p < 0.001)
could be identified as independent predictive risk factors.
Only low birth weight (0.27 (0.11–0.66), p = 0.004) could
be identified as a single protective factor, while no combina-
tion of protective factors was significant.
Conclusion: The significance and the informative value of
risk and protective factor combinations, e.g., for selective ul-
trasound surveys, are limited. Early universal ultrasound
screening using the Graf method is advised to timely detect
hip immaturity and pathologies and to provide the optimal
approach for mature or immature but appropriate for age and
pathologic hips.
What is Known:
• Ultrasound screening is essential for diagnosis and treatment of DDH in
time while the weighting and correlation of protective factors and
perinatal risk and their role for the ultrasound screening protocol still
give rise to debate.
What is New:
• The effects of single risk and protective factors for DDH do not cumulate
or counteract, resulting in a limited value of a selective screening
protocol based on risk and protective factors.
• A universal screening protocol using the Graf ultrasound method is
recommended.
Keywords Regression analysis . Odds ratio . Sonographic
neonatal screening . Hip dislocation . Congenital
Abbreviations
DDH Developmental dysplasia of the hip

Introduction
Developmental dysplasia of the hip (DDH) is one of the most
frequent congenital abnormalities in newborns [56, 60] and
may lead to serious impairment if diagnosed and treated too
late [30, 39, 60]. Due to the application and improvement of
imaging procedures for the examination of the newborns’ hip
joints, the term DDH has been broadened from being merely a
clinical diagnosis of evidently unstable and (sub)luxated hips
to more subtle distinctions and maturity grades [20, 26, 27].
An established method to evaluate the appropriate maturation
of the newborn’s hip joints and to prevent DDH is the ultra-
sound hip screening method invented by Graf in 1980 [12, 16,
17]. The subsequent implementation in several countries led
to less late and severe DDH presentations [26, 30, 57, 60].
However, numerous new questions regarding the survey and
treatment constellations arose from the usage of the Graf ul-
trasound method. Major points of discussion remain the reli-
ability of the method [46, 61], the implementation of a univer-
sal or selective ultrasound screening program [46], the optimal
timing of the ultrasound investigations [27, 57], the diagnostic
relationship to clinical tests [9, 28], and the identification,
weighting, and correlation of perinatal risk and protective fac-
tors to estimate the hip maturation and the probability for
DDH development, especially when suggested as a precondi-
tion before a selective ultrasound screening policy [22, 34, 38,
41]. The aim of this retrospective data analysis was to corre-
late single and twofold combined risk and protective factors
for DDH with the hip maturity, objectively evaluated with the
Graf ultrasound method within the first 2 to 5 days of life.
Patients and methods
All neonates born in the Klinik Hirslanden, Zurich, generally
undergo a clinical and ultrasound examination of the hips
according to the Graf ultrasound method within the first 2 to
5 days (to avoid missing out newborns in a later appointment)
[29] after delivery in the Department of Neonatology. All
ultrasound examinations are performed by mere one experi-
enced ultrasound examiner (MS) who is certified in hip so-
nography, according to a standardized examination and ther-
apy protocol. The condition of the hip joints is determined
according to the Graf method. Further (therapeutic) steps de-
pend on the clinical and ultrasound findings in the first and the
follow-up examinations.
Data assessment
From January 1993 to May 2011, 11,922 neonates were
born in the Klinik Hirslanden, Zurich. For analysis, only
Eur J Pediatr
complete data sets with entire anamnesis, risk and protec-
tive factor assessment, and ultrasound data were used.
Newborns with musculoskeletal and neurogenic impair-
ments were excluded. All data sets were anonymized be-
fore statistical analysis. Accordingly, an ethic approval was
not necessary for this research setting at that time.
Assessed risk and protective factors (yes/no) were female
gender, high birth weight, breech presentation at birth, posi-
tive family history for DDH, prematurity (34th to 37th gesta-
tional week), low birth weight, and twin birth. Relative birth
weight was differentiated between the 10th and 90th percen-
tile as normal for gestational age, below as small, and above as
large for gestational age, respectively [33, 62].
Graf method
The ultrasound assessment of the hip according to Graf quan-
tifies the maturity of the cartilaginous and bony acetabular
roof and the position of the femoral head based on sonograph-
ic structures [12–14, 16, 17] (Fig. 1). For hip examination, a
newborn positioning device and an ultrasound probe guide
system (both Gebrueder Hirschbeck GmbH, St. Peter am
Kammersberg, Austria) were used (Fig. 2). All examinations
were performed by the same neonatologist (MS), using a 7.5-
MHz linear transducer (Siemens Sonoline SI-250 and G-20,
Berlin and Munich, Germany). The hip type and the following
steps are based on the ultrasound findings of the poorer hip
[11, 17, 45, 52]:
Graf type I: no treatment and re-examination after 4 to
6 weeks.
Graf type IIa: instruction of the neonate’s parents for
common conservative steps providing a physiologic ab-
duction (double diaper, rare swaddling, supine position)
and re-examination after 4 to 6 weeks. The parents are
thoroughly informed about these steps and involved in
the procedure to maximize the compliance and the
outcome.
Graf type IIc, D, III, IV: immediate orthopedic conserva-
tive treatment (usually with Pavlik harness or Tuebinger
splint), performed in an affiliated pediatric orthopedic
hospital, with monthly clinical and ultrasound follow-up
until documented successful treatment, and additional x-
ray of the hips at the onset of walking, upon school entry
and at the onset of adolescence.
Graf type IIb: the hip joints did not reach type I after
3 months and remain type IIa. As this type is diagnosed
at a later time point, it is not included in our data analysis
of the first ultrasound examination within the first 2 to
5 days of life.
Eur J Pediatr
Fig. 1 Hip types according to Graf (type I (a), IIa (b), IIc (c), D (d), III
(e), and IV (f)). For a valid hip sonogram, the anatomical orientation is
crucial and the following three key landmarks must be identified to
evaluate the standard plane for the examination: the lower limb of the
os ilium (1), the mid section of the bony acetabular roof (2), and the
Statistical analysis
For statistical analysis, two groups were built to form a yes/no
variable. Graf ultrasound types I and IIa were considered ma-
ture or immature but appropriate for age and merged into the
first group. Type IIc, D, III, and IV were considered patholog-
ic and potentially leading to DDH and merged into the second
group. In a first step, univariate analyses using Pearson’s chi-
squared test and Fisher’s exact test (the latter when the expect-
ed count was <5) were performed to evaluate the impact of
Fig. 2 Ultrasound examination of the newborn’s hip according to Graf.
The newborn is placed in the positioning device (1), the ultrasound probe
(2) is fixed in the guide system (3). The examiner operates the ultrasound
probe with his left hand (4) and guides it additionally with his right hand
(5) while the newborn’s mother or the nurse additionally fixes the
newborn with both hands (6)
acetabular labrum (3). Additionally, the femoral head (number sign), the
chondro-osseous junction (plus sign), the synovial fold (asterisk), the
bony rim (section sign), the hyaline cartilage performing the acetabular
roof (percent sign), and the joint capsule (ampersand) have been marked
in the physiological hip sonogram (type I)
single and twofold combined potential risk and protective fac-
tors (i.e., female gender, high birth weight, breech presenta-
tion at birth, positive family history, prematurity, low birth
weight, and twin birth, yes/no variable) on the presence of
an ultrasonographic pathological hip type (i.e., IIc, D, III, IV,
yes/no variable). To avoid very small sample rates, we
refrained from analyzing multiple factor combinations or a
multinominal logistic regression model to analyze the impact
on detailed single pathological hip types. In a second step, risk
and protective factors that were found to be statistically sig-
nificant with a p value of <0.05 were entered into a logistic
regression analysis to evaluate the unbiased impact and pre-
dictive value of single and twofold combined risk and protec-
tive factors (independent variables) on the presence of a ultra-
sonographic pathological hip type (dependent variable). For
final model selection, the backward elimination method was
used to select the factors that contributed to explain the hip
pathology. p values <0.05 were considered significant. All
statistical analyses were performed with SPSS® Statistics
(Version 22 for Windows, Armonk, NY, USA).
Results
Complete data sets were available for 11,820 of all 11,922
universally screened newborns (99.1%). Risk and protective
factor distribution and detailed ultrasonographic hip types ac-
cording to the Graf method are given in Table 1. Univariate
analyses for single and twofold combined risk and protective
factors associated with pathologic hip types are given in
Table 2.
 Eur J Pediatr

The final logistic regression model for the impact of single
risk and single protective factors on the pathological hip types
is given in Table 3. None of the entered risk or protective
factors (which matched significance in the univariate analysis)
was eliminated within the modeling. The final model ex-
plained 12.9% (Nagelkerke R square) of the variance in hip
pathology and correctly classified 53.1% of all cases (overall
classification percentage). Pathologic hip types could be cor-
rectly identified in 86.6% of the cases; mature or immature but
appropriate for age hip types could be correctly identified in
52.3% of the cases.
The final logistic regression model for the impact of single
and twofold risk and protective factor combinations on path-
ologic hips is given in Table 4. This model explained 13.4% of
the variance in hip pathology and correctly classified 53% of
the cases. Pathologic hip types could be correctly identified in
87.3% of the cases; mature or immature but appropriate for
age hip types could be correctly identified in 52.1% of the
cases. High birth weight, female gender and breech presenta-
tion, female gender and positive family history, high birth
weight and breech presentation, high birth weight, and posi-
tive family anamnesis did not reach significant levels and were
eliminated by the logistic regression model.
From all 11,820 followed-up and reviewed newborns,
surgical reduction was necessary in one case (0.008%).
Here, follow-up investigations of the female newborn
(39th week of pregnancy, birth weight 3295 g, no family
Table 2 Univariate analysis for single and combined risk and
protective factors associated with pathological ultrasonographic hip
types according to the Graf method
p value

Female gender <0.001*

Table 1 Risk and protective factor distribution and detailed High birth weight <0.001*
ultrasonographic diagnoses Breech presentation <0.001*
Positive family history <0.001*
Mature or immature Pathologic (n, %)
but appropriate for age (n, %) Prematurity 0.11
Low birth weight 0.002*
Total 11,529 (97.5) 291 (2.5) Twin birth 0.002*
Gender Female gender
Male 6076 (52.7) 56 (19.2) And high birth weight <0.001*
Female 5453 (47.3) 235 (80.8) And breech presentation <0.001*
Diagnosis And positive family history <0.001*
I 8663 (75.1) 0 (0) And prematurity 0.77
IIa 2866 (24.9) 0 (0) And low birth weight 0.18
IIc 0 (0) 112 (38.5) And twin birth 0.18
D 0 (0) 112 (38.5) High birth weight
III 0 (0) 63 (21.7) And breech presentation <0.001*
IV 0 (0) 4 (1.3) And positive family history <0.001*
Birth weight And prematurity 0.39
Low 628 (5.4) 5 (1.72) And twin birth >0.99
Normal 9889 (85.8) 237 (81.44) Breech presentation
High 1012 (8.8) 49 (16.84) And positive family history 0.18
Breech presentation And prematurity 0.07
Yes 800 (6.9) 67 (23) And low birth weight 0.77
No 10,729 (93.1) 224 (77) And twin birth 0.77
Family history Positive family history
Yes 368 (3.2) 39 (13.4) And prematurity 0.33
No 11,161 (96.8) 252 (86.6) And low birth weight 0.31
Prematurity And twin birth >0.999
Yes 501 (4.35) 7 (2.4) Prematurity
No 11,028 (95.65) 284 (97.6) And low birth weight 0.38
Twin birth And twin birth 0.016*
Yes 450 (3.9) 2 (0.7)
No 11,079 (96.1) 289 (99.3) For analysis, Pearson’s chi-squared test and Fisher’s exact test (when the
expected count was <5) were used. Single and combined factors matching
All values are given as total numbers (percentage); ultrasonographic hip significant levels were used for logistic regression analysis
types are presented according to the Graf ultrasound method *A p value <0.05 was considered significant
Eur J Pediatr

Table 3 Logistic regression

modeling with backward Parameter estimate β OR (95% CI) p value
elimination for single risk and
protective factors correlated with Female gender 1.62 5.05 (3.74–6.81) <0.001*
pathological hip types (IIb, D, III, High birth weight 0.97 2.63 (1.8–3.65) <0.001*
IV) Breech presentation 1.58 4.86 (3.62–6.54) <0.001*
Positive family history 1.6 4.93 (3.42–7.13) <0.001*
Low birth weight −1.33 0.27 (0.11–0.66) 0.004*
Twin birth −1.9 0.15 (0.04–0.62) 0.009*

OR odds ratio, 95% CI 95% confidence interval

*A p value <0.05 was considered significant

history for DDH, left hip joint initially type III) were
missed and the young patient underwent abundant diag-
nostics for abdominal pain in an external institution, which
finally turned out to originate from a severe hip dysplasia
on the left side.
Discussion
In our study, the distribution of pathological hip types assessed
with the Graf method is comparable to that of other studies
[26, 27, 49, 59, 60]. We could confirm female gender [1, 2, 9,
11, 27, 42, 49, 59, 63], high birth weight [2, 11], breech pre-
sentation at birth [9, 11, 23, 27, 38, 42], and positive family
history [9, 11, 27, 38, 42] as independent risk factors as well as
a low birth weight [27] and twin birth [7, 8, 10, 43, 45] as
independent protective factors for DDH in a logistic regres-
sion model as they were significantly correlated with
pathological hip types (Table 3). When considered in twofold
combinations, several combinations were significantly more
common in newborns with pathological hip types (Table 2),
but only the combination of female gender with high birth
weight showed an independent predictive value for DDH in
our logistic regression model (Table 4). Additionally, high
birth weight and twin birth failed to show an impact on the
ultrasonographic hip type if considered within twofold risk
and protective factor combinations. In accordance with other
authors, we could not identify prematurity as a risk factor for
DDH [9, 47]. Because it would have resulted in very small
sample rates, we refrained from analyzing triple factor combi-
nations or a multinominal logistic regression model to analyze
the impact on detailed single pathological hip types.
Since the introduction of different hip ultrasound examina-
tion methods, mainly by Graf in 1980 [12], Harcke in 1984 [4,
21], Suzuki in 1987 [50], and Terjesen in 1988 [53–55], the
early diagnosis of pathologic hip joints and consecutive DDH

Table 4 Logistic regression

modeling with backward Parameter estimate β OR (95% CI) p value
elimination for single and
combined risk and protective Female gender 1.4 4.07 (3.01–5.51) <0.001*
factors correlated with Breech presentation 1.61 4.98 (3.71–6.71) <0.001*
pathological hip types (IIb, D, III, Positive family anamnesis 1.62 5.05 (3.49–7.31) <0.001*
IV)
Low birth weight −1.32 0.27 (0.11–0.66) 0.004*
Twin birth −1.38 0.25 (0.06–1.05) 0.059
Female gender and high birth weight 1.26 3.51 (2.45–5.03) <0.001*
Prematurity and twin birth −16.16 <0.001 >0.99
High birth weight −0.23 0.8 (0.32–2.01) 0.63
Female gender and breech presentation 0.07 1.07 (0.52–2.23) 0.85
Female gender and positive family history −0.09 0.91 (0.39–2.14) 0.83
High birth weight and breech presentation 0.23 1.26 (0.46–3.45) 0.66
High birth weight and positive family anamnesis 0.37 1.45 (0.5–4.15) 0.49

Prematurity combined with twin birth did not reach significant levels, but was not eliminated by the logistic
regression modeling. The given result is a statistical artifact due to the fact that in the subgroup of premature twin
newborns, no pathologic hips occurred. From a modeling point of view, this situation leads to computational
limitations as illustrated by the logistic regression results
OR odds ratio, 95% CI 95% confidence interval
*A p value <0.05 was considered significant

has improved essentially, while late and severe onsets of DDH
showed a significant decrease compared to the time where
only clinical tests were performed [26, 30, 36, 52, 57, 60].
While the general benefit from an ultrasound screening of
the hip, e.g., like performed by national screening programs
in Austria (1991), Switzerland (1995), and Germany (1996),
using the Graf method) [17] is undeniable, the favored ultra-
sound method, as well as the screening approach (universal
vs. selective), is still under debate [36, 40].
Regarding the ultrasound method of choice, a main argu-
ment against a screening with the Graf method is its suscepti-
bility to false-positive results and the consecutive risk for
overtreatment [59, 61] and its observer dependence [48, 61].
Our data analysis shows that the maturity grades of the Graf
ultrasound method show a highly significant correlation with
predisposing risk factors for DDH if the investigations are
performed by an experienced sonographer who is certified in
performing hip sonography and the possibility for interobserv-
er variations and bias is eliminated. With a universal ultra-
sound screening according to the Graf method, followed by
a three-stage protocol consisting of (1) no treatment, (2)
common conservative steps providing a physiologic abduction
(each with an ultrasonographic re-examination after 4 to
6 weeks), and (3) orthopedic conservative treatment (with a
Pavlik harness or a Tuebinger splint with continuous clinical
and radiologic surveillance), the progressive orthopedic treat-
ment and its risk for adverse effects, like avascular femoral
head necrosis [60], remain restricted only to persistent patho-
logic hip types. At the same time, not only evidently patho-
logical and (sub)luxated, but also immature but appropriate
for age hip joints, which may resolve spontaneously under
non-orthopedic conservative steps, can be detected and com-
mon conservative steps can be initiated [11, 41, 59]. Here, the
highly standardized, reliable, reproducible, and safe Graf
method with objective and quantitative measurement param-
eters like the alpha and beta angle for the evaluation of bony
and cartilaginous roof has a crucial advantage compared to the
Harcke [4, 5, 21], the Suzuki [36, 51], or the Terjesen method
[36, 54, 55], which do not discriminate anatomic details of the
newborns’ hip joints [4, 11, 37, 41, 59, 60]. Consequently,
expert knowledge in combination with continuing education
of the examiner remains the main premise for a low rate of
over- and underdiagnosis and overtreatment while being able
to detect subtle maturity grades using the Graf ultrasound
method [1, 26, 27, 39, 46, 48, 58, 59, 61].
Regarding the screening protocol (universal vs. selective),
the selection of newborns based on the presence of single or
combined risk factors has been proposed for a selective ultra-
sound screening with the aim of reducing overdiagnosis and
overtreatment [3, 22, 34, 38, 40, 41] and ultimately avoiding
unnecessary health care costs [31, 41]. However, subjecting
the ultrasound examination of the hip to perinatal anamnestic
risk factors leads to certain inconsistent shortcomings. First,
Eur J Pediatr
several authors demonstrated that if only Bat-risk^ neonates
and infants would have been screened in their survey, only
half [11, 15] or less [2, 6] of the cases who required conser-
vative steps or orthopedic treatment due to immature or path-
ologic hip joints would have been detected and treated in time.
Kohler et al. argue that if risk factors are present, ultrasound
does not constitute a screening method anymore, but an inves-
tigation to clarify the joint conditions and the diagnosis [25].
Moreover, if all risk factors that have been described (e.g.,
additional oligohydramnios, maternal hypertension, maternal
age, type of birth, previous miscarriage, social class, birth
rank) [2, 3] would be considered within the selection for ul-
trasound screening, only a small group (e.g., boys without
anamnestic risk factors) would remain uninvestigated [2].
Furthermore, numerous studies that investigated selective
screening programs based on single risk factors failed to detect
a significant decrease in open surgical reductions [24, 37, 60]
or a higher cost-effectiveness compared to a universal screen-
ing protocol [6, 18, 19, 56, 60].
Finally, our data suggests that single and twofold combined
risk factors may contribute to the occurrence of DDH as they
occur significantly more often in newborns with immature and
pathological hips. However, apart from the combination of
female gender with high birth weight, no risk factor combina-
tion reached a significant level in the logistic regression mod-
el. With regard to the low overall explained variance of the
model, the predictive value of risk factor combinations is lim-
ited and may not lead to a more convincing selection for
selective ultrasound investigations, e.g., by detecting minor
and major risk groups with one or more risk factors.
With the present method, i.e., universal ultrasound screen-
ing using the Graf ultrasound method within the first 2 to
5 days of life, only one open reduction had to be performed
over a span of 18 years in 11,820 cases, while all other con-
ditions developed properly under conservative approaches.
This rate of 0.008% is comparable to other findings of early
universal screening surveys. Tschauner et al. reported a rate of
zero open reductions, if only universally screened newborns
were considered [60]; Clegg et al. reported no late presenta-
tion of DDH, and only 0.006% of all cases required open
surgery after failed conservative treatment [6]. As well,
Bache et al. have had no case of late DDH since universal
ultrasound screening (here using the Harcke method) has been
established [2]. Consequently, we join the opinion that an
early universal ultrasound screening is required to timely de-
tect hip immaturity and pathologies and to provide the optimal
conservative approach for mature or immature but appropriate
for age and pathologic hips in time [6, 27, 32, 35, 44, 57, 60].
As suggested in recent publications, this may even lead to a
higher cost-effectiveness than a selective screening [6, 18, 19,
56, 60]. If universal screening is not possible to be performed
by a certified sonographer during the first week of life due to
human or infrastructural resources, compromising universal
Eur J Pediatr
protocols, e.g., like proposed by Thallinger et al. (early screen-
ing for newborns with risk factors, later screening for new-
borns without risk factors), should be applied [57].
Summarizing and concluding, we cannot recommend se-
lective ultrasound screening based on single risk factors like it
has been proposed in the past [22, 34, 38, 40, 41]. Also, the
combination of two risk or risk and protective factors does not
lead to significantly higher evidence about minor and major
risk groups for DDH. Therefore, we cannot support selective
ultrasound screening protocols based on an alleged accumu-
lated impact of risk factor combinations, and hence, we rec-
ommend a universal screening protocol. Further evidence is
needed for the optimal timing of the ultrasound examinations
and the development of different hip types during the first
weeks of life.
Authors’ contribution The first draft of the manuscript has been writ-
ten by Dr. Labruyère and corrected by Dr. Schams. The statistic admin-
istration, further creation, and finalization were performed by Dr. Walensi
and Dr. Zuse. The photos were shot by Dr. Schams; the revision of the
manuscript was performed by Dr. Walensi, Dr. Schams, and Dr.
Labruyère.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflicts of
interest.
Ethical approval This article does not contain any studies with human
participants or animals performed by any of the authors.
References
1. Arti H, Mehdinasab SA, Arti S (2013) Comparing results of clinical
versus ultrasonographic examination in developmental dysplasia of
hip. J Res Med Sci 18:1051–1055
2. Bache CE, Clegg J, Herron M (2002) Risk factors for developmen-
tal dysplasia of the hip: ultrasonographic findings in the neonatal
period. J Pediatr Orthop B 11:212–218
3. Chan A, McCaul KA, Cundy PJ, Haan EA, Byron-Scott R (1997)
Perinatal risk factors for developmental dysplasia of the hip. Arch
Dis Child Fetal Neonatal Ed 76:F94–100
4. Clarke NM, Harcke HT, McHugh P, Lee MS, Borns PF, MacEwen
GD (1985) Real-time ultrasound in the diagnosis of congenital dis-
location and dysplasia of the hip. J Bone Joint Surg Br 67:406–412
5. Clarke NM, Clegg J, Al-Chalabi AN (1989) Ultrasound screening
of hips at risk for CDH. Failure to reduce the incidence of late cases.
J Bone Joint Surg Br 71:9–12
6. Clegg J, Bache CE, Raut VV (1999) Financial justification for
routine ultrasound screening of the neonatal hip. J Bone Joint
Surg Br 81:852–857
7. Czubak J, Mazela JL, Majda W, Woźniak W (2003) Is twin preg-
nancy a risk factor for devolopmental dysplasia of the hip - retro-
spective analysis using ultrasonography. Orthop Traumatol Rehabil
5:712–716
8. De Pellegrin M, Moharamzadeh D (2010) Developmental dysplasia
of the hip in twins: the importance of mechanical factors in the
etiology of DDH. J Pediatr Orthop 30:774–778
9. Dogruel H, Atalar H, Yavuz OY, Sayli U (2008) Clinical examina-
tion versus ultrasonography in detecting developmental dysplasia
of the hip. Int Orthop 32:415–419
10. Doyle PE, Beral V, Botting B, Wale CJ (1991) Congenital
malformations in twins in England and Wales. J Epidemiol
Community Health 45:43–48
11. Falliner A, Hahne HJ, Hassenpflug J (1999) Sonographic hip
screening and early management of developmental dysplasia of
the hip. J Pediatr Orthop B 8:112–117
12. Graf R (1980) The diagnosis of congenital hip-joint dislocation by
the ultrasonic Combound treatment. Arch Orthop Trauma Surg 97:
117–133
13. Graf R (1981) The ultrasonic image of the acetabular rim in infants.
An experimental and clinical investigation. Arch Orthop Trauma
Surg 99:35–41
14. Graf R (1983) New possibilities for the diagnosis of congenital hip
joint dislocation by ultrasonography. J Pediatr Orthop 3:354–359
15. Graf R (1997) Hip ultrasonography. Basic principles and current
aspects. Orthopade 26:14–24
16. Graf R (2006) Hip sonography—diagnosis and management of
infant hip Dysplasia, 2nd edn. Springer Verlag, Berlin
17. Graf R, Mohajer M, Plattner F (2013) Hip sonography update.
Quality-management, catastrophes—tips and tricks. Med Ultrason
15:299–303
18. Gray A, Elbourne D, Dezateux C, King A, Quinn A, Gardner F
(2005) Economic evaluation of ultrasonography in the diagnosis
and management of developmental hip dysplasia in the United
Kingdom and Ireland. J Bone Joint Surg Am 87:2472–2479
19. Grill F, Müller D (1997) Results of hip ultrasonographic screening
in Austria. Orthopade 26:25–32
20. Gulati V, Eseonu K, Sayani J, Ismail N, Uzoigwe C, Choudhury
MZ, Gulati P, Aqil A, Tibrewal S (2013) Developmental dysplasia
of the hip in the newborn: a systematic review. World J Orthop 4:
32–41
21. Harcke HT, Clarke NM, Lee MS, Borns PF, MacEwen GD (1981)
Examination of the infant hip with real-time ultrasonography. J
Ultrasound Med 3:131–137
22. Holen KJ, Tegnander A, Bredland T, Johansen OJ, Saether OD,
Eik-Nes SH, Terjesen T (2002) Universal or selective screening of
the neonatal hip using ultrasound? A prospective, randomised trial
of 15,529 newborn infants. J Bone Joint Surg Br 84:886–890
23. Imrie M, Scott V, Stearns P, Bastrom T, Mubarak SJ (2010) Is
ultrasound screening for DDH in babies born breech sufficient? J
Child Orthop 4:3–8
24. Kamath S, Mehdi A, Wilson N, Duncan R (2007) The lack of
evidence of the effect of selective ultrasound screening on the inci-
dence of late developmental dysplasia of the hip in the Greater
Glasgow Region. J Pediatr Orthop B 16:189–191
25. Kohler G, Hell AK (2003) Experiences in diagnosis and treatment
of hip dislocation and dysplasia in populations screened by the
ultrasound method of Graf. Swiss Med Wkly 133:484–487
26. Kokavec M, Bialik V (2007) Developmental dysplasia of the hip.
Prevention and real incidence. Bratisl Lek Listy 108:251–254
27. Kolb A, Schweiger N, Mailath-Pokorny M, Kaider A, Hobusch G,
Chiari C, Windhager R (2013) Low incidence of early developmen-
tal dysplasia of the hip in universal ultrasonographic screening of
newborns: analysis and evaluation of risk factors. Int Orthop 40:
123–127
28. Koşar P, Ergun E, Unlübay D, Koşar U (2009) Comparison of
morphologic and dynamic US methods in examination of the new-
born hip. Diagn Interv Radiol 15:284–289
29. Köse N, Omeroğlu H, Ozyurt B, Akçar N, Ozçelik A, Inan U, Seber
S (2006) Our three-year experience with an ultrasonographic hip
screening program conducted in infants at 3 to 4 weeks of age. Acta
Orthop Traumatol Turc 40:285–290

30. von Kries R, Ihme N, Oberle D, Lorani A, Stark R, Altenhofen L,
Niethard FU (2003) Effect of ultrasound screening on the rate of
first operative procedures for developmental hip dysplasia in
Germany. Lancet 362:1883–1887
31. Lee J (2008) Developmental dysplasia of the hip: universal or se-
lective ultrasound screening? Ann Acad Med Singap 37:101–103
32. Lewis K, Jones DA, Powell N (1999) Ultrasound and neonatal hip
screening: the five-year results of a prospective study in high-risk
babies. J Pediatr Orthop 19:760–762
33. Lubchenco LO, Hansman C, Dressler M, Boyd E (1963)
Intrauterine growth as estimated from liveborn birth-weight data
at 24 to 42 weeks of gestation. Pediatrics 32:793–800
34. Mahan ST, Katz JN, Kim YJ (2009) To screen or not to screen? A
decision analysis of the utility of screening for developmental dys-
plasia of the hip. J Bone Joint Surg Am 91:1705–1719
35. Marks DS, Clegg J, al-Chalabi AN (1994) Routine ultrasound
screening for neonatal hip instability. Can it abolish late-
presenting congenital dislocation of the hip? J Bone Joint Surg Br
76:534–538
36. Omeroğlu H (2014) Use of ultrasonography in developmental dys-
plasia of the hip. J Child Orthop 8:105–113
37. Paton RW, Hossain S, Eccles K (2002) Eight-year prospective
targeted ultrasound screening program for instability and at-risk
hip joints in developmental dysplasia of the hip. J Pediatr Orthop
22:338–341
38. Paton RW, Hinduja K, Thomas CD (2005) The significance of at-
risk factors in ultrasound surveillance of developmental dysplasia
of the hip. A ten-year prospective study. J Bone Joint Surg Br 87:
1264–1266
39. Riccabona M, Schweintzger G, Grill F, Graf R (2013) Screening for
developmental hip dysplasia (DDH)-clinically or sonographically?
Comments to the current discussion and proposals. Pediatr Radiol
43:637–640
40. Rosendahl K, Toma P (2007) Ultrasound in the diagnosis of devel-
opmental dysplasia of the hip in newborns. The European approach.
A review of methods, accuracy and clinical validity. Eur Radiol 7:
1960–1967
41. Rosendahl K, Markestad T, Lie RT, Sudmann E, Geitung JT (1995)
Cost-effectiveness of alternative screening strategies for develop-
mental dysplasia of the hip. Arch Pediatr Adolesc Med 149:643–
648
42. Rühmann O, Konermann W, Lazović D, Vitek L, Bouklas P (1998)
Ultrasound neonatal screening: the effect of anamnestic risk factors
on hip dysplasia. Z Orthop Ihre Grenzgeb 136:492–500
43. Rühmann O, Lazović D, Bouklas P, Rössig S (1998) Ultrasound hip
joint screening in newborn infants. Is twin pregnancy a risk factor
for dysplasia? Ultraschall Med 19:64–69
44. Rühmann O, Lazović D, Bouklas P, Gossé F, Franke J (1999)
Ultrasound hip joint screening in newborn infants. Correlation of
anamnestic risk factors and hip dysplasia. Klin Padiatr 211:141–
148
45. Schwend RM, Schoenecker P, Richards BS, Flynn JM, Vitale M
(2007) Pediatric Orthopaedic Society of North America. Screening
the newborn for developmental dysplasia of the hip: now what do
we do? J Pediatr Orthop 27:607–610
46. Sewell MD, Eastwood DM (2011) Screening and treatment in de-
velopmental dysplasia of the hip-where do we go from here? Int
Orthop 35:1359–1367
Eur J Pediatr
47. Sezer C, Unlu S, Demirkale I, Altay M, Kapicioglu S, Bozkurt M
(2013) Prevalence of developmental dysplasia of the hip in preterm
infants with maternal risk factors. J Child Orthop 7:257–261
48. Simon EA, Saur F, Buerge M, Glaab R, Roos M, Kohler G (2004)
Inter-observer agreement of ultrasonographic measurement of alpha
and beta angles and the final type classification based on the Graf
method. Swiss Med Wkly 134:671–677
49. Sionek A, Czubak J, Kornacka M, Grabowski B (2008) Evaluation
of risk factors in developmental dysplasia of the hip in children
from multiple pregnancies: results of hip ultrasonography using
Graf's method. Orthop Traumatol Rehabil 10:115–130
50. Suzuki S, Awaya G, Wakita S, Maekawa M, Ikeda T (1987)
Diagnosis by ultrasound of congenital dislocation of the hip joint.
Clin Orthop Relat Res 217:171–178
51. Suzuki S, Kasahara Y, Futami T, Ushikubo S, Tsuchiya T (1991)
Ultrasonography in congenital dislocation of the hip. Simultaneous
imaging of both hips from in front. J Bone Joint Surg Br 73:879–
883
52. Synder M, Harcke HT, Domzalski M (2006) Role of ultrasound in
the diagnosis and management of developmental dysplasia of the
hip: an international perspective. Orthop Clin N Am 37:141–147
53. Terjesen T, Bredland T, Eik-Nes SH (1988) Ultrasonic examination
of hip joints in newborn infants. Tidsskr Nor Laegeforen 108:2359–
2363
54. Terjesen T, Bredland T, Berg V (1989) Ultrasound for hip assess-
ment in the newborn. J Bone Joint Surg Br 71:767–773
55. Terjesen T, Rundén TO, Tangerud A (1989) Ultrasonography and
radiography of the hip in infants. Acta Orthop Scand 60:651–660
56. Thaler M, Biedermann R, Lair J, Krismer M, Landauer F (2011)
Cost-effectiveness of universal ultrasound screening compared with
clinical examination alone in the diagnosis and treatment of neona-
tal hip dysplasia in Austria. J Bone Joint Surg Br 93:1126–1130
57. Thallinger C, Pospischill R, Ganger R, Radler C, Krall C, Grill F
(2014) Long-term results of a nationwide general ultrasound
screening system for developmental disorders of the hip: the
Austrian hip screening program. J Child Orthop 8:3–10
58. Tönnis D, Storch K, Ulbrich H (1990) Results of newborn screen-
ing for CDH with and without sonography and correlation of risk
factors. J Pediatr Orthop 10:145–152
59. Treiber M, Tomazic T, Tekauc-Golob A, Zolger J, Korpar B, Burja
S, Takac I, Sikosek A (2008) Ultrasound screening for developmen-
tal dysplasia of the hip in the newborn: a population-based study in
the Maribor region, 1997-2005. Wien Klin Wochenschr 120:31–36
60. Tschauner C, Fürntrath F, Saba Y, Berghold A, Radl R (2011)
Developmental dysplasia of the hip: impact of sonographic new-
born hip screening on the outcome of early treated decentered hip
joints-a single center retrospective comparative cohort study based
on Graf's method of hip ultrasonography. J Child Orthop 5:415–424
61. Vasilescu D, Cosma D, Vasilescu DE, Botar Jid C, Gersak M,
Lenghel M, Dudea SM (2015) A new sign in the standard hip
ultrasound image of the Graf method. Med Ultrason 17:206–210
62. Voigt M, Schneider KT, Jährig K (1996) Analysis of a 1992 birth
sample in Germany. 1: new percentile values of the body weight of
newborn infants. Geburtshilfe Frauenheilkd 56:550–558
63. Yiv BC, Saidin R, Cundy PJ, Tgetgel JD, Aguilar J, McCaul KA,
Keane RJ, Chan A, Scott H (1997) Developmental dysplasia of the
hip in South Australia in 1991: prevalence and risk factors. J
Paediatr Child Health 33:151–156