Ultrasonography in screening for developmental dysplasia of the hip in newborns: systematic reviewBMJ 2005; 330 doi: https://doi.org/10.1136/bmj.38450.646088.E0 (Published 16 June 2005) Cite this as: BMJ 2005;330:1413
All rapid responses
October 10th, 2005
In BMJ,doi:10.1136/bmj.38450.646088.EO (published 1 June 2005),
Woolacott et al. presents a systematic review on ultrasonography in
screening for developmental dysplasia (DDH) of the hip in newborns. The
authors have done an extensive amount of work based on data from the
existing medical literature; however, a correct use of such data requires
a minimum of knowledge about the diagnostic methods used, and about the
Unfortunately, the paper is flawed with inconsistencies. First, hip
ultrasound was developed by Graf alone, who started his experimental work
in 1978, thoroughly described the sonoanatomy of the hip joint, and
defined the standard section through the deepest part of the acetabulum,
upon which most modified techniques are based on1-21. In 1984, Harcke
reported on a dynamic technique while Morin in 1985 used the “femoral head
coverage” in classifying the hips into normal and pathological 22;23. His
technique was slightly modified by Terjesen in 198924.
Second, although the authors aimed at examining papers on newborns,
three out of ten target articles address infant screening (20, 25, 26).
One of their ten target articles, ref.28, does not appear in Tables 1 and
2, which is rather confusing. Moreover, the authors wrongly state that a
modified technique after Terjesen was used in our randomised trial (ref
21), although they state that we chose to treat new-borns presenting with
Graf type IIIa or worse25. The correct information is that we used Graf’s
technique, with a simplified classification into morphologically normal,
immature, slight dysplasia or severe dysplasia according to a-angle, +/-
Third, the authors give the readers the impression that DDH in new-
borns include hip instability alone, which subsequently develops into
dysplasia in cases of persisting subluxation or dislocation, and that
dysplasia diagnosed in newborns is one entity. This is not the case. After
refining the ultrasound techniques and classification systems used, it has
become possible to classify dysplasia into severe or mild, with or without
additional instability, and to tailor treatment and follow-up thereafter.
Postponing therapy for those suffering severe dysplasia/dislocation would
result in late DDH with dislocation for a high per cent.
Differing between varying degrees of late DDH when discussing the
effect of screening strategies is crucial. When comparing the rates of
late DDH between the two RCT’s included in the present review, the authors
did not discuss the fact that the criteria for late DDH between one and
four months of age, when the diagnosis was based on ultrasound, differed
between the two studies. Moreover, they suggested that the differences in
rates might reflect an increased level of experience since the most
recently published paper reported on lower rates. However, data-collection
was performed during the same period of time (1988-1990/92). The different
rates are more likely to be caused by differences in ultrasound technique
used (modified Morin vs. modified Graf), the number of examiners (one-
examiner study vs. four examiners) and the criteria used for early and for
Since the pathology in most cases of DDH is already present at birth,
the term late DDH is commonly defined as DDH diagnosed after one month of
age. In the present review, the authors argue that the term late DDH
should be used after several months, when the diagnosis is more obvious,
and when those showing physiologically immaturity in the new-born period
have normalized. They also recommend that a randomized study comparing the
effect of ultrasound screening at one and at three months of age should be
performed. I do not agree. In my opinion a major challenge for future work
on hip-screening is to standardise the ultrasound techniques used, to
agree on criteria for early and late DDH and to ensure that the examiners
are adequately trained.
(1) Graf R. The diagnosis of congenital hip-joint dislocation by the
ultrasonic Combound treatment. Arch Orthop Trauma Surg 1980; 97(2):117-
(2) Graf R. The ultrasonic image of the acetabular rim in infants.
An experimental and clinical investigation. Arch Orthop Trauma Surg 1981;
(3) Graf R. [The anatomical structures of the infantile hip and its
sonographic representation]. Morphol Med 1982; 2(1):29-38.
(4) Graf R. [The importance of sonography in the examination of the
infant hip]. Biomed Tech (Berl) 1983; 28(11):257-263.
(5) Graf R. New possibilities for the diagnosis of congenital hip
joint dislocation by ultrasonography. J Pediatr Orthop 1983; 3(3):354-359.
(6) Graf R. [The sonographic evaluation of hip dysplasia using
convexity diagnosis]. Z Orthop Ihre Grenzgeb 1983; 121(6):693-702.
(7) Graf R. Fundamentals of sonographic diagnosis of infant hip
dysplasia. J Pediatr Orthop 1984; 4(6):735-740.
(8) Graf R. Classification of hip joint dysplasia by means of
sonography. Arch Orthop Trauma Surg 1984; 102(4):248-255.
(9) Graf R, Heuberer I. [Sonographically monitored hip dysplasia].
Wien Klin Wochenschr 1985; 97(1):18-27.
(10) Graf R. [Hip sonography in infancy. Procedure and clinical
significance]. Fortschr Med 1985; 103(4):62-66.
(11) Graf R. [Possibilities, problems and present status of hip
sonography of infant hips]. Radiologe 1985; 25(3):127-134.
(12) Graf R, Tschauner C, Steindl M. [Does the IIa hip need
treatment? Results of a longitudinal study of sonographically controlled
hips of infants less than 3 months of age]. Monatsschr Kinderheilkd 1987;
(13) Graf R. [Sonographic diagnosis of hip dysplasia. Principles,
sources of error and consequences]. Ultraschall Med 1987; 8(1):2-8.
(14) Graf R. [Sonography of the hip in infants]. Z Orthop Ihre
Grenzgeb 1990; 128(4):355-356.
(15) Graf R. Hip sonography--how reliable? Sector scanning versus
linear scanning? Dynamic versus static examination? Clin Orthop Relat Res
(16) Graf R. [Hip ultrasonography. Basic principles and current
aspects]. Orthopade 1997; 26(1):14-24.
(17) Graf R. [Developmental hip disorders in infants. Diagnosis and
therapy]. Orthopade 1997; 26(1):1.
(18) Graf R. Ultrasound measurements of the newborn hip--comparison
of two methods in 657 newborns. Acta Orthop Scand 1998; 69(5):550-551.
(19) Graf R. [Ultrasound examination of the hip. An update].
Orthopade 2002; 31(2):181-189.
(20) Tschauner C. Earliest diagnosis of congenital dislocation of
the hip by ultrasonography. Historical background and present state of
Graf's method. Acta Orthop Belg 1990; 56(1 ( Pt A)):65-77.
(21) Tschauner C, Klapsch W, Graf R. [The effect of ultrasonography
screening of hips in newborn infants on femur head necrosis and the rate
of surgical interventions]. Orthopade 1993; 22(5):268-276.
(22) Harcke HT, Clarke NM, Lee MS, Borns PF, MacEwen GD. Examination
of the infant hip with real-time ultrasonography. J Ultrasound Med 1984;
(23) Morin C, Harcke HT, MacEwen GD. The infant hip: real-time US
assessment of acetabular development. Radiology 1985; 157(3):673-677.
(24) Terjesen T, Bredland T, Berg V. Ultrasound for hip assessment
in the newborn. J Bone Joint Surg Br 1989; 71(5):767-773.
(25) 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(1):47-52.
(26) Rosendahl K, Markestad T, Lie RT. Developmental dysplasia of
the hip: prevalence based on ultrasound diagnosis. Pediatr Radiol 1996;
Competing interests: No competing interests
We read with interest the review article by Woolacott et al.
´Ultrasonography in screening for developmental dysplasia of the hip in
newborns: systematic review´(10). It was an extensive and complete review
including articles that are not in English. However we would have to
disagree with the final conclusion and there are numerous points we would
like to highlight:
Firstly in many of the studies including the one from Rosendahl et
al. in 1994 infants were scanned early in the first week of life (8).
Timing for ultrasound is crucial. The majority of the babies scanned this
early will have a Type IIa (Graf) hip and as a consequence will require
further screening and possibly postural treatment (although not proven to
be effective). This will increase costs and anxiety. However ultrasound
scanning at 60 days of life has shown to reduce the number of subjects
with type IIa hip by 64 – 78 % (1). This could help reduce overall costs,
especially in view of most of these hips being normal at follow-up (1).
The second point we would like to mention is that in this report the
only study commenting about sensitivity in ultrasound revealed a value of
88.5 % (7). On the other hand clinical examination sensitivities of 87 –
99 % in highly experienced examiners have been reported (1,5). Comparing
the accuracy of clinical examination to ultrasound screening seems
reasonable for a study, but does not reflect common practice. While in
England hip examinations are mainly performed by the least experienced
member of staff (SHO) at discharge from hospital, the national ultrasound
screening at 4 - 8 weeks of age in Germany is only performed by registered
primary care paediatricians after a compulsory training of at least 400
supervised scanned hips. The paediatric training for general practioners
in the UK is also quite variable, so that you cannot commonly assume a
high experience, either. Moreover sensitivity and positive predictive
value of the Barlow-Ortolani test are extremely low if sonography was
assumed to be the gold standard and if sonography results worse than type
IIa were considered pathologic (1). There might not be enough clear
evidence in favour of or against general ultrasound screening, but there
is certainly not much evidence in favour of clinical examination, either
Thirdly cost-effectiveness was not addressed, probably because
current evidence is controversial. However in a recent cost-effectiveness
analysis from the Netherlands general ultrasound screening at 3 months did
not seem to be more expensive than clinical examination (6). Reports from
the UK Hip Trial and from Rosendahl in 1995 came to similar conclusions in
the past (3,9). It would be probably expensive to exchange a clinical hip
screening with an ultrasound screening system depending on the existing
health care system and training system of the physicians involved. This
question has not been investigated, yet.
So, if general ultrasound screening was to be introduced in the UK,
who would do it? Paediatricians, GPs, Orthopaedics or Radiologists? How
many more doctors would you need to meet the requirements in the
population? What would be the training like in order to achieve a
nationwide consistently high expertise with this ultrasound test? Is the
current clinical training sufficient to say it is as good as ultrasound
screening? Is it worthwhile going through all these questions if the
difference is probably going to be marginal?
Our conclusion is that depending on the health care system and the
training system of physicians general ultrasound screening of the hips is
probably the more sensitive, more cost effective and less intrusive method
for the patient´s routine screening examination of the hips. It helps to
reduce late DDH and is associated with shorter and less intrusive
1. Baronciani D, Atti G, Andiloro F et al. Screening for
developmental dysplasia of the hip: From theory to practice. Pediatrics
1997; 99(2): E5.
2. Bialik V. Clinical hip instability in the newborn by an orthopaedic
surgeon and a paediatrician. J Pediatr Orthop 1986; 6: 703 – 705.
3. Elbourne D, Dezateux C, Arthur R et al. Ultrasonography in the
diagnosis and management of developmental hip dysplasia (UK hip trial)
:clinical and economic results of a multicentre randomised controlled
trial. Lancet 2002; 360: 2009 – 2017.
4. Godward S, Dezateux C. Surgery for congenital dislocation of the hip in
the UK as a measure of outcome of screening. MRC working party on CDH.
Lancet 1998; 351 (9110): 1149 – 1152.
5. Patel H and Canadian Task Force on Preventive Health Care. Preventive
health care, 2001 update: Screening and management of developmental
dysplasia of the hip in newborns. CMAJ 2001; 164 (12): 1669 – 1677.
6. Roovers EA, Boere-Boonekamp MM, Zielhuis GA et al. Post-neonatal
ultrasound screening for developmental dysplasia of the hip. A study of
cost-effectiveness in the Netherlands [doctoral thesis]. Eschede,
Netherlands: University of Twente, 2004.
7. Roovers EA, Boere-Boonekamp MM, Castelein RM et al. Effectiveness of
ultrasound screening for developmental dysplasia of the hip. Arch Dis
Child Fetal Neonatal Ed 2005; 90: F25 – 30.
8. 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.
9. Rosendahl K. Cost-effectiveness of alternative screening strategies for
developmental dysplasia of the hip. Arch Ped Adolesc Med 1995; 149: 643-
10. Woolacott NF, Puhan MA, Steurer J and Kleijnen J. Ultrasonography in
screening for developmental dysplasia of the hip in newborns: systematic
review. BMJ 2005; 330: 1413 – 1418.
Competing interests: No competing interests