Abstract In this prospective study, the role of clinicalexamination and risk factors in the diagnosis of develop-mental dysplasia of the hip (DDH) was analysed. For thispurpose, 376 hips of 188 referred infants with an averageage of 3.7 months (range 0–8 months) were clinically andultrasonographically (Graf method) examined. Infantswho had at least one risk factor for DDH had a three timeshigher occurence rate of DDH than those who didn’t haveany. Breech presentation and positive family history werethe two most common risk factors associated with DDH.The rate of DDH was nearly 16 times higher in the hipswith at least one abnormal clinical finding than in the oneswithout any. Ultrasonographically unstable hips (type IIc-unstable to type IV) were always detected during clinicalexamination. There was a risk of missing the diagnosis ofhips with acetabular dysplasia alone. Limitation of abduc-tion and asymmetry of the skin folds were the two mostcommon clinical findings associated with DDH. Nearlytwo-thirds of the infants with at least one positive clinicalfinding and at least one risk factor had either an unilateralor bilateral dysplastic hip. It was concluded that a carefulclinical examination was significantly more importantthan the coexisting risk factors in the detection of DDH.

Introduction

Early detection of developmental dysplasia of the hip(DDH) is known to be the most important step for obtain-ing better clinical and radiological outcomes. Currently,there are two widely used methods for the early detectionof developmental dysplasia of the hip: clinical screeningand ultrasound screening. Clinical examination plays aconsiderable role in the diagnosis of DDH. Ortolani andBarlow’s tests are the two important clinical findings frombirth to 2 months of age. At 3–12 months of age, limita-tion of abduction of the hip, Galeazzi sign, asymmetry ofthe thigh and inguinal skin folds, and telescoping signsare important clinical findings [17]. Prior to performingthese tests, some factors commonly associated with DDHsuch as positive family history, breech presentation, fe-male sex, oligohydramnios, and associated congenital ano-malies should be considered [17].

It is currently believed that clinical examination hasnot been successful in finding all children with DDH [10].Hip ultrasonography is a sensitive and useful techniquefor both the early diagnosis and treatment of DDH, and atpresent it is the best method of assessment up to 6 monthsof age [6, 7, 17].

The purpose of this study was to assess the role of clin-ical examination and risk factors in the diagnosis of DDHin young infants. Hip ultrasonography was used as the fi-nal diagnostic method.

Patients and methods

This was a prospective study which was planned to end within 6 months, and it included 188 young infants (99 girls and 89 boys)who were referred to the first author (experienced in the field ofpaediatric orthopaedics and especially in the diagnosis and treat-ment of DDH). Infants with hip dislocations due to teratologiccauses or neuromuscular disorders were not included in the study.Their age ranged from 4 days to 8 months with an average age of3.7 months. Before performing the clinical examination, all knownanamnestic and existing (associated anomalies) risk factors forDDH were questioned and documented. Clinical examinations of

Hakan Ömeroglu · Süha Koparal

The role of clinical examination and risk factors in the diagnosis of developmental dysplasia of the hip: a prospective study in 188 referred young infants

Arch Orthop Trauma Surg (2001) 121 :7–11 © Springer-Verlag 2001

Received: 29 September 1999

ORIGINAL ARTICLE

A part of this study was presented as a free paper at the 2nd German-Turkish Congress of the Pediatric Orthopaedics,Izmir, Turkey, May 29-June 1, 1998

H. ÖmerogluOsmangazi University Hospital, Department of Orthopaedics and Traumatology, 26480 Eskișehir, Turkey

S. KoparalAnkara Numune Training and Research Hospital, 3rd Clinic of Radiology, 06100 Ankara, Turkey

H. Ömeroglu (�)Turgut Reis Caddesi 54/8, 06570 Ankara, Turkeye-mail: homer@ogu.edu.tr

all infants were performed by the first author. Clinical examinationwas always performed when the infant was completely free of anx-iety, and forceful manipulations were always avoided. All clinicalfindings were recorded in detail. After completion of the clinicalexamination, the second author (experienced in the field of paedi-atric hip radiology) carried out ultrasonography of both of the in-fant’s hips with the Graf technique [6]. Ultrasonographic examina-tions of all hips were performed in the lateral decubitis positionwith the same ultrasound equipment (Hitachi EUB-420, 64 mmlinear 7.5/5-mHz transducer). At least two print-outs of the sono-graphic images were obtained to avoid ultrasonographic misdiag-nosis. Definite ultrasonographic typing of the hips according to theGraf classification system, which had an adequate level of reliabil-ity and reproducibility [14], was made following a careful assesse-ment of the sonographic print-outs by two authors leading to com-plete agreement as to the type of hip. We prefer to have a dedicatedorthopaedist-radiologist examination team for the detection ofDDH as this kind of team study produces a more qualified exami-nation and is more convenient for the infant’s parents and thephysicians [4].

Statistical analysis of the data was done by using a PC program(SPSS 7.5 for Windows). The chi-square test and Kolmogorov-Smirnov test for two groups were used. A p value < 0.05 was con-sidered significant.

Results

According to the Graf classification system, type I andtype IIa+ hips are normal and need no treatment. Type IIa–

to type IV hips are abnormal and require conservative orsurgical treatment [6]. DDH was observed in 32 of 188 in-fants (17%). Among the 376 hips examined, 43 (11%)were considered ultrasonographically abnormal, and theinvolved side was bilateral in 11, right in 8 and left in 13cases. (Table 1). Unilateral or bilateral DDH was seen in22 of 99 (22%) girls and in 10 of 89 (11%) boys. The rateof occurrence of DDH (abnormal hip ultrasonography)was not statistically different between girls and boys (chi-square test, p = 0.07).

Of the 188 infants, 85 had (45%) at least one risk fac-tor for DDH. The rate of occurrence of unilateral or bilat-eral DDH was 27% in infants who had at least one riskfactor for DDH and 9% in infants who did not have any(Fig.1). This difference was statistically significant (chi-square test, p = 0.002). A positive family history andbreech presentation were the two most common factorsassociated with DDH (Table 2). The overall rate of posi-tive family history was 33/188 (18%) and breech presen-

tation, 12/188 (6%). The rate of occurrence of DDH was11/33 (33%) in infants with a positive family history and7/12 (58%) in ones with a breech presentation. The in-volved side was bilateral in 5 of 11 (45%) infants with apositive family history and in 2 of 7 (29%) in infants withbreech presentation. The ultrasonographic hip type wasconsidered unstable (types IIc-unstable to IV) in 12 of 16 hips (75%) with a positive family history and in 4 of 8 hips (50%) with a breech presentation. An abnormal ul-trasonographic finding was seen in 1 of 6 infants whowere twins and in 2 of 5 infants with oligohydramnios.When coexisting congenital anomalies were taken intoconsideration, DDH was not detected in 16 infants withtalipes equinovarus nor in 4 with torticollis but was de-tected in 2 of 5 infants with pes calcaneovalgus and in 1 of 2 infants with metatarsus varus.

Clinical findings except asymmetry of the skinfoldswere recorded for each hip. Asymmetry of the skinfoldswas assessed separately as they were recorded for each in-fant, whereas other clinical findings were recorded foreach hip. Limitation of abduction was described as a hipabduction of less than 70 deg [16, 22]. The rate of occur-

8

Table 1 Ultrasonographic hiptypes of 376 examined hips according to the Graf classifi-cation

Hip type (Graf) No.

Type I 329Type IIa+ 4Type IIa– 3Type IIb 13Type IIc-stable 2Type IIc-unstable 3Type D 11Type IIIa 6Type IV 5

Total 376

Fig.1 Correlation between risk factors for developmental dyspla-sia of the hip (DDH) and ultrasonographic findings in 188 younginfants

Table 2 Associated risk factors in infants with DDH

Risk factor No.

Positive family history 10Breech presentation 5Breech presentation + pes calcaneovalgus 2Oligohydramnios 1Positive family history + oligohydramnios 1Tibial agenesis 1Twin pregnancy 1Congenital spine deformity 1Metatarsus varus 1No risk factors 9

Total 32

rence of DDH was 47% in the hips with at least one ab-normal clinical finding and 3% in the ones without any(Fig.2). The difference was considered very significant(chi-square test, p < 0.00001). Limitation of abductionwas the most common clinical finding in the 33 clinicallyand ultrasonographically abnormal hips (Table 3). Among333 ultrasonographically normal hips, limitation of ab-duction was seen in 36 (11%). Of ten hips with normalclinical but abnormal ultrasonographic examinations, 9 wereultrasonographically immature or critical hips (3 type IIa–,5 type IIb and 1 type IIc- stable) and 1 was ultrasono-graphically unstable (type D). Among 9 ultrasonographi-cally abnormal hips of 6 infants younger than 8 weeks, 1was type IIc-stable, 1 type D, 3 type IIIa and 4 type IV. In3 of those 9 hips (1 type D and 2 type IIIa), the Ortolanitest was positive. In all type IV hips, limitation of abduc-tion was the only positive clinical finding. All type III andIV hips, 10 of 11 type D hips, and all type IIc-unstablehips had at least one positive clinical finding when asym-metry of the skinfolds was not taken into consideration.

Asymmetry of the skinfolds was seen in 45 (24%) in-fants. The rate of unilateral or bilateral abnormal hip ul-trasonography was 10% and 38% in infants without andwith skinfold asymmetry, respectively (Fig.3). This dif-ference was significant (chi-square test, p = 0.00006). In 5 unilateral cases (1 type IIa–, 2 type IIb, 1 type IIc and 1 type D), the only positive finding was the asymmetry of

the skinfolds. Among 17 infants with both DDH andasymmetry of the skinfolds, the involved side was bilat-eral in 7 and unilateral in 10 infants. In seven bilateralcases at least one side was always ultrasonographicallyunstable.

The rate of occurance of unilateral or bilateral DDHwas 4% in infants with a negative clinical finding in bothhips (including the skinfold asymmetry) and negative riskfactors, 2% in infants with negative clinical finding inboth hips and positive risk factors, 18% in infants with atleast one positive clinical finding in one or both hips andnegative risk factors, and 65% in infants with at least onepositive clinical finding in one or both hips and positiverisk factors (Fig. 4). The difference between the four groupswas very significant (Kolmogorov-Smirnov test for twogroups, p < 0.00001). The ultrasonographic type of abnor-mal hips belonged to the first two groups was 3 unilateralIIb and 1 bilateral IIa–.

9

Fig.2 Correlation between clinical findings (excluding skinfoldasymmetry) and ultrasonographic findings in 376 hips

Fig.3 Correlation between asymmetry of the skinfolds and ultra-sonographic findings in 188 infants

Fig.4 Correlation between DDH and risk factors and clinicalfindings in infants. (Cl clinical examination)

Table 3 Clinical findings in hips with DDH

Clinical finding No.

Limitation of abduction 24Limitation of abduction + Galeazzi 5Ortolani 2Limitation of abduction + Ortolani 1Galeazzi 1Normal clinical examination 10

Total 43

Discussion

The findings of our study revealed that the rate of DDHwas nearly 16 times higher in the hips which had at leastone abnormal clinical finding (except skinfold asymme-try) than in the ones which did not have any. It was alsoevident that an experienced physician could always detectdecentric, subluxated and dislocated hips during a clinicalexamination, but he/she could miss hips with only acetab-ular dysplasia. It seems that hip ultrasonography plays amore important role in the early detection of dysplastichips without any evident clinical finding.

Tönnis et al. [22] reported that more than half of thetype IIc and D hips had no palpatory findings, but nearly80% of the type III and 100% of the type IV hips had pal-patory findings. The findings about type IIc and type Dhips do not definitely coincide with ours. Falliner et al. [5]found 63% of the clinically normal hips to be ‘sonographi-cally dysplastic’. In the present study, we observed thatnearly half of the hips with abnormal clinical finding, ex-cluding the asymmetry of the skinfolds, were dysplastic,and in infants younger than 8 weeks the Ortolani test andlimitation of abduction and in ones older than 8 weekslimitation of abduction were the most common clinicalfindings associated with DDH. Limitation of abductionwas not only the most common positive clinical findingbut also the most common reason for referral as in someother studies involving the detection of DDH by ultra-sonography [1, 16, 20, 22]. Limitation of abduction wasseen in approximately one-tenth of normal hips, and thismight be due to anxiety of the infant during clinical ex-amination, even though optimal conditions had been ob-tained. Also, racial variations may influence the results.We suggest that limitation of abduction is an importantclinical finding but is not always associated with DDH.

An ultrasonographic examination was previouslystated as appropriate for confirmation of a positive Or-tolani test [11]. In another study, it was suggesed that thepresence of a positive Ortolani sign was associated with aseverer sonographic pathology than that associated with apositive Barlow’s test in unstable hips [1]. Hip clicks andBarlow’s test are already considered less important clini-cal findings for the detection of DDH [2, 13]. Our find-ings seem to support this.

We found asymmetry of the skinfolds to be an impor-tant clinical finding associated with DDH for all agegroups. The risk of DDH was nearly four times higher inthe infants with asymmetry of the skinfolds. This asym-metry has been suggested to be more marked in unilateralcases [17], and we have seen a slight predominance forunilateral cases. We can also say that if asymmetry of theskinfolds is observed in bilateral cases, at least one hip isusually expected to be unstable.

It has been stated that regardless of the approach to ex-amination, hip ultrasonography is universally accepted asbeing more sensitive than clinical examination [9]. Teg-nander et al. [19] suggested that a clinical examination byan experienced doctor was superior to examination by

someone with less experience. Not examining an infanttoo frequently and too forcibly and performing a gentleexamination increase the success rate and benefits of clin-ical examination [12]. Failure of the clinical examinationlies with the examiner and not with the tests [8]. We havecompletely obeyed these rules and obtained a satisfactorycorrelation between clinical examination and hip ultra-sonography.

Our study confirmed the findings of some previousstudies that two risk factors (breech presentation and apositive family history) were frequently associated withabnormal hip ultrasonography [3, 5, 8, 11, 13, 16, 18, 22].In this study, unilateral or bilateral DDH was observedthree times more often in infants who had at least one riskfactor for DDH than in ones who didn’t have any. DDHwas detected in nearly two-thirds of the infants with abreech presentation and in one-third of the infants with apositive family history. More than half of the involvedhips were considered unstable. The presence of these tworisk factors may influence the severity of the hip dyspla-sia. We believe that the correlation between some risk fac-tors and the severity of hip dysplasia has to be analysed ina larger series. We also observed that pes calcaneovalgusand metatarsus varus were the two lower limb anomalieswhich should also alert the physician for detection ofDDH.

Paton et al. [15] stated that ultrasound screening ofclinically unstable hips alone or associated with at-riskfactors detected DDH at a high rate, but ultrasoundscreening of at-risk hips on their own was of little value.The findings of our study coincide well with theirs.

The incidence of DDH was 17% in the present study.This is not a true incidence in the general population as allpatients were referred to the first author from other physi-cians to eliminate or confirm the diagnosis of DDH. DDHhas a definite predilection for girls, and the ratios reportedin the international literature range between 2.4:1 and9.2:1. The involved side is usually unilateral, and the lefthip is affected more frequently. Definite racial and geo-graphic variations may influence these ratios [17, 21]. Al-though the number of infants examined in the presentstudy was not large enough to make accurate commentson the sex and involved side, our ratios coincide with theclassic values in the textbooks.

In conclusion, positive clinical findings from a carefulclinical examination is significantly more important thanthe existence of at-risk factors in the diagnosis of DDH,and hips with abnormal clinical findings must always beexamined ultrasonographically. The highest suspicion ofDDH is reserved for infants with positive clinical findingsand with positive risk factors. However, it must be re-membered that a normal clinical examination does not al-ways prove the absence of DDH.

References

1.Bialik V, Wiener F (1993) Sonography of suspected develop-mental dysplasia of the hip: a description of 3624 hips. J Pedi-atr Orthop Part B 2: 152–155

10

2.Bond CD, Hennrikus WL, DellaMaggiore ED (1997) Prospec-tive evaluation of newborn soft-tissue hip clicks with ultra-sound. J Pediatr Orthop 17: 199–201

3.Castelein RM, Sauter AJM, Vlieger M de, Linge B van (1992)Natural history of ultrasound hip abnormalities in clinicallynormal newborns. J Pediatr Orthop 12: 423–427

4.Davids JR, Benson LJ, Mubarak SJ, McNeil N (1995) Ultra-sonography and developmental dysplasia of the hip: a cost-benefit analysis of three delivery systems. J Pediatr Orthop 15:325–329

5.Falliner A, Hahne HJ, Hassenpflug J (1999) Sonographic hipscreening and early management of developmental dysplasia ofthe hip. J Pediatr Orthop Part B 8: 112–117

6.Graf R, Wilson B (1995) Sonography of the infant hip and itstherapeutic implications. Chapman&amp;Hall, Weinheim

7.Grissom LE, Harcke HT (1999) Ultrasonography and develop-mental dysplasia of the infant hip. Curr Opin Pediatr 11: 66–69

8.Hadlow V (1988) Neonatal screening for congenital dislocationof the hip; a prospective 21-year survey. J Bone Joint Surg Br70: 740–743

9.Harcke HT, Kumar JK (1991) The role of ultrasound in the di-agnosis and management of congenital dislocation and dyspla-sia of the hip. J Bone Joint Surg Am 73: 622–628

10.Hensinger RN (1995) The changing role of ultrasound in themanagement of developmental dysplasia of the hip (DDH). J Pediatr Orthop 15: 723–724

11.Holen KJ, Terjesen T, Tegnander A, Bredland T, Saether OD,Eik-Nes SH (1994) Ultrasound screening for hip dysplasia innewborns. J Pediatr Orthop 14: 667–673

12.Moore FH (1989) Examining infant hips: can it do harm? J Bone Joint Surg Br 71: 4–5

13.Nimityongskul P, Hudgens RA, Anderson LD, Melhem RE,Green AE Jr, Saleeb SF (1995) Ultrasonography in the man-agement of developmental dysplasia of the hip (DDH). J Pedi-atr Orthop 15: 741–746

14.Ömeroglu H, Biçimoglu A (1999) A multivariance analysis ofhip ultrasonography by the Graf method in developmental dys-plasia of the hip. 18th Meeting of the European Paediatric Or-thopaedic Society, Göteborg, Sweden

15.Paton RW, Srinivasan MS, Shah B, Hollis S (1999) Ultrasoundscreening for hips at risk in developmental dysplasia. Is it worthit? J Bone Joint Surg Br 81: 255–258

16.Stoffelen D, Urlus M, Molanaers G, Fabry G (1995) Ultra-sound, radiographs and clinical symptoms in developmentaldislocation of the hip: a study of 170 patients. J Pediatr OrthopPart B 4: 194–199

17.Tachdjian MO (1997) Clinical pediatric orthopedics; the art ofdiagnosis and principles of management. Appleton&Lange,Stamford

18.Teanby DN, Paton RW (1997) Ultrasound screening for con-genital dislocation of the hip: a limited targeted programme. J Pediatr Orthop 17: 202–204

19.Tegnander A, Terjesen T, Bredland T, Holen KJ (1994) Inci-dence of late-diagnosed hip dysplasia after different screeningmethods in newborns. J Pediatr Orthop Part B 3: 86–88

20.Terjesen T (1996) Ultrasound as the primary imaging methodin the diagnosis of hip dysplasia in children aged < 2 years. J Pediatr Orthop Part B 5: 123–128

21.Tönnis D (1987) Congenital dysplasia and dislocation of thehip in children and adults. Springer, Berlin Heidelberg NewYork

22.Tönnis D, Storch K, Ulbrich H (1990) Results of newbornscreening for CDH with and without sonography and correla-tion of risk factors. J Pediatr Orthop 10: 145–152

11