The Journal : Current Issue


	The Journal		Protocol Reviews		eResearch Archive		Info for Authors		Profile


Help		Feedback

The Journal

	Current Issue



	Content= s=20 in full
	T= alking=20 points
	Early=20 online publication
	Original=20 research
	News
	Editorial=20 and review
	C= orrespondence
	Us= es=20 of error
	Clinical=20 picture
	= Dissecting=20 room
	Department=20 of error




	Supplements

	Back Issues



		Search Journal

		Subscribe

		E-Toc

		Log Out


		Jobs
		Awards = and=20 Announcements
		Conferences
		Press Services
		General Info
		Free Sample
		Contact Us

Home

The Journal

Current Issue<= /FONT>

Original=20 research

Volume 362, Number 9399 =20 06 December 2003


	Articles

Effect of = ultrasound=20 screening on the rate of first operative procedures = for=20 developmental hip dysplasia in Germany=20

R=FCdiger von Kries, Nicola Ihme, Doris Oberle, = Anette=20 Lorani, Renee Stark, Lutz Altenhofen, Fritz Uwe Niethard =

Department for Paediatric Epidemiology, Institute for = Social=20 Paediatrics and Adolescent Medicine, Ludwig-Maximilians = University,=20 Munich, Germany (Prof R von Kries MD, D Oberle PhD, R = Stark=20 MD); Orthopaedic Department, University Aachen (N = Ihme MD, A=20 Lorani MD, Prof F Uwe Niethard MD); and Central Research=20 Institute of Ambulatory Health Care in Germany, Cologne = (L=20 Altenhofen PhD)

Correspondence to: Prof R=FCdiger von Kries, = Department for=20 Paediatric Epidemiology, Institute for Social Paediatrics = and=20 Adolescent Medicine, Ludwig-Maximilians University, = Heiglhofstr 63,=20 81377 Munich, Germany (e-mail:AG.EPI@LRZ.uni-muenchen.de= )

Summary=20
Introduction=20
Methods=20
Results=20
Discussion=20
References=20

Summary =

Background Ultrasound screening for = developmental hip=20 dysplasia and early conservative treatment might prevent = later hip=20 operations. A national hip ultrasound-screening programme,=20 undertaken during the first 6 weeks of life, was introduced = across=20 Germany in 1996 and was continued for 5 years. We assessed = the rate=20 of first operation on the hip during this programme.=20

Methods A national active surveillance = programme of=20 initial operations for developmental hip dysplasia was = started in=20 1997 and continued for 5 years. Screening participation was = assessed=20 by a random digit dialling telephone survey. Cases were = children=20 aged between 10 weeks and 5 years at first operation, who = had had no=20 underlying disease leading to developmental hip dysplasia.=20 Completeness of case ascertainment was validated with a=20 capture-recapture study in a representative subsample. = Calculated=20 incidences were compared with previously established rates.=20

Findings About 90% of all children were = screened. 147=20 cases in the first year, and between 81 and 105 for = subsequent years=20 were reported. Treatment included closed reductions 353 = (66%), open=20 reductions 61 (11%), and osteotomies 121 (23%). = Developmental hip=20 dysplasia was diagnosed by ultrasound before 6 weeks of age = in 272=20 (55%) of cases, 64 (13%) were screened at the recommended = time but=20 had normal findings, 70 (14%) had delayed screening, and 89 = (18%)=20 were not screened. Capture-recapture estimates suggested = that 52% of=20 cases were reported. The corrected incidence for first = operation was=20 0=B726 per 1000 livebirths (95% CI 0=B722-0=B732).=20

Interpretation Ultrasound screening seems to = prevent=20 many, but not all, operations for developmental hip = dysplasia. Rates=20 of timely screening (ie, before 6 weeks of age) and training = of=20 doctors in ultrasound screening need to be improved.=20

Lancet 2003; 362: 1883-87

Introduction =

Developmental hip dysplasia can lead to impaired hip = function=20 and premature degenerative joint disease because of higher = than=20 normal load and shearing forces.¹ Diagnosis of = this=20 abnormality within the first weeks of life is essential for = early=20 abduction treatment, which aims to prevent a deleterious = course,=20 leading to hip disease.^2,3 Universal screening = based on=20 clinical signs has therefore been recommended for more than = half a=20 century.⁴

A report by Godward and Dezateux⁵ showed = that the=20 clinical screening programme for developmental hip dysplasia = in the=20 UK was not universally successful in preventing operative = procedures=20 related to the disease. The researchers challenged the = present=20 screening programme that is based on clinical signs and = symptoms,=20 and called for an assessment of alternative screening = policies,=20 including universal ultrasound imaging.⁵ Since = the=20 numbers of children with the disease who need an operation = is small,=20 randomised trials assessing the effectiveness of ultrasound=20 screening on the rate of such procedures would need to = assess large=20 numbers of children to achieve significant results. Valid = survey=20 data might provide an alternative.=20

In Germany, a screening programme for developmental = hip=20 dysplasia that included universal static⁶ = ultrasound=20 imaging for all children was started in January, = 1996.⁷=20 All children are screened clinically for the abnormality in = the=20 first week of life. Children with a family history, breech = delivery,=20 or suspicious findings after clinical examination are = screened with=20 ultrasound in the first week of life. Screening for all = other=20 children is scheduled at age 4-6 weeks--the latest at end of = week 6.=20

Active surveillance for operations for this condition = in=20 children younger than 5 years was initiated in 1997 and = maintained=20 until 2002. The goal of this active surveillance was to = establish=20 whether the rates of first operation for developmental hip = dysplasia=20 were lower with a universal ultrasound screening programme = compared=20 with historical data in Germany⁸ and other=20 countries,^9,10 where screening was based only on = clinical=20 signs and symptoms. A further goal was to see which, if any, = operative procedures became necessary despite early = diagnosis by the=20 ultrasound screening programme.

Methods =

Cases who had had operations were identified through = active=20 surveillance from May, 1997 to April, 2002, by use of the = German=20 paediatric surveillance unit (ESPED).¹¹ Monthly = postcards=20 were sent to all heads of orthopaedic hospitals or = departments=20 caring for children younger than 5 years. They were to = report the=20 number of cases admitted for operative procedures in the = previous=20 month, or had the option of nothing to report. Reported = cases were=20 validated by a follow-up questionnaire for country of birth, = underlying disease (such as arthrogryposis, = myelomeningocele, or=20 cerebral palsy), date of ultrasound screening, ultrasound = findings=20 ( 3Dalpha=20

angle, type according to Graf⁶), type = of=20 operation done during the reported admission, and any other = such=20 procedures previously done on the same child.=20

Eligibility criteria for cases were: born in Germany, = aged=20 older than 9 weeks and younger than 5 years with a first = operation=20 in the month of surveillance, and no underlying disease = leading to=20 developmental hip dysplasia or acquired dislocations. = Operative=20 procedures included closed reduction, open reduction, and = osteotomy=20 with or without reduction, in a child admitted to hospital. = Closed=20 reduction could mean any of the following: arthrography, = adductor=20 tenotomy, traction treatment, plaster-of-Paris casts, = harness, or a=20 combination of these. All questionnaires were validated by = an=20 orthopaedic surgeon.=20

Completeness of the case ascertainment was assessed = with the=20 capture-recapture method in southern Bavaria,¹² = where=20 representatives from all orthopaedic hospitals were asked to = give=20 permission for an onsite visit. The investigator attempted = to=20 identify all children younger than 5 years who had had an = operative=20 procedure for the disease between May, 1997 and April, 2001, = in the=20 respective hospitals from international classification of = disease=20 coding lists, or operative or admission records. These cases = constituted the independent second data source, and were = linked to=20 those from the same hospitals identified by the active = surveillance=20 programme.=20

Cases with first operations in every year of the = surveillance=20 study could have been born up to 5 years before the year in = which=20 the outcome of interest was assessed. Thus, estimation of = true=20 incidence for defined birth cohorts was impossible. British=20 investigators⁵ reported incidence estimates as = the number=20 of events recorded during the observation period related to = the=20 number of births during this period. We estimated incidence=20 accordingly--first operative procedures for every year of=20 surveillance by the respective numbers of livebirths in = 1997, 1998,=20 1999, 2000, and 2001.=20

Screening programme participation was assessed in a=20 population of children identified by random digit dialling. = Parents=20 were asked whether their children had had ultrasound = screening of=20 the hip, and, if the answer were yes, they were asked when = this=20 examination had been done. The answer options were during = your stay=20 in the maternity unit, during the third wellbaby check-up = test=20 between the 4th and 6th week of life, or other time. In a = survey=20 undertaken in 1999, roughly 25000 randomly selected = households were=20 asked whether they had children aged younger than 3 years. = If they=20 had, they were invited to participate in a vaccination = coverage=20 survey.¹³ Families with children born between = 1996 and=20 1998 were contacted again in 2002 by telephone or mail to = answer the=20 ultrasound screening questions. For the 1999 to 2001 = cohorts, a=20 similar random digit dialling survey was undertaken in 2002 = to=20 assess vaccination coverage. Questions about ultrasound = screening=20 were included in this survey, and families were under no = obligation=20 to participate. We obtained approval from the Bavarian = office for=20 protection of data confidentiality. Since our investigation = used=20 anonymous data and did not interfere with treatment = decisions, we=20 did not contact any of the patients with developmental hip = dysplasia=20 or their parents. The data were entered in an Access data = bank.=20 Descriptive statistics were calculated in SAS, version 6.12. =

Role of the funding source=20

The funding source had no role in study design, data=20 collection, data analysis, data interpretation, or in the = writing of=20 the report.

Results =

The proportion of children who were screened by = ultrasound=20 increased from 83% in 1996 to 95% or above in most = subsequent years=20 (figure). Yearly, between 26% and 45% children had = ultrasound=20 screening for developmental hip dysplasia in the first week = of life,=20 with no trend over time.=20

=20

Proportion of children screened by = ultrasound for developmental hip = dysplasia=20

Vertical bars are 95% CI. =

The orthopaedic surgeons returned the cards on which = cases=20 were reported and follow-up questionnaires validating the = cases in=20 1997-2002, at a rate that ranged from 89% to 98% (cards) and = from 90=20 to 98% (questionnaires) per year. Many of the 1887 returned=20 questionnaires were for children who did not meet criteria = for=20 inclusion--608 were younger than 10 weeks, 27 were older = than 5=20 years, 102 had had no operative procedure according to the = case=20 definition, 36 were not treated in hospital, 91 had not been = born in=20 Germany, 134 had either acquired or teratological = developmental hip=20 dysplasia, 112 had had previous operations, and 33 were = treated=20 before the study period. Additionally, there were 181 double = reports=20 and 28 reports without essential information, which left 535 = children with first operative procedures for final analyses. = The=20 yearly number of first procedures reported ranged between 81 = and 147=20 cases per year (table 1) with the highest number of first = operative=20 procedures recorded in 1997. This finding was mainly due to = a higher=20 number of closed reductions in 1997 than in subsequent = years. In 54=20 of the 121 osteotomies, a reduction of the hip was also = done.=20

	May, 1997-April, 1998	May, 1998-April, 1999	May, 1999-April, 2000	May, 2000-April, 2001	May, 2001-April, 2002	Total
Closed reduction	91 (62%)	73 (70%)	65 (64%)	60 (74%)	64 (63%)	353 (66%)
Open reduction	20 (14%)	14 (13%)	14 (14%)	6 (7%)	7 (7%)	61 (11%)
Osteotomy	36 (25%)	18 (17%)	22 (22%)	15 (19%)	30 (30%)	121 (23%)
Total	147 (100%)	105 (100%)	101 (100%)	81 (100%)	101 (100%)	535 (100%)
Data are number (%).
Table 1: Children with first = operative=20 procedures by year of = observation

Closed reductions accounted for two-thirds of all = operative=20 procedures (table 1). Of the 353 children treated with = closed=20 reduction, 216 were treated by fixation with a = plaster-of-Paris cast=20 or a harness, 166 had traction treatment, 95 had = arthrography, and=20 13 had adductor tenotomy. 236 (67%) children had only one = procedure=20 for closed reduction, 97 (28%) had two, and 20 (6%) had = three=20 operative procedures.=20

For 495 of the 535 cases, time of ultrasound = screening and=20 screening result were known (table 2). For 259 of the 272 = children=20 with a first operative procedure and an abnormal screening=20 ultrasound within the recommended timeframe, the date of the = final=20 diagnosis of was reported: 133 (51%) were 28 days old or = younger and=20 227 (88%) were 90 days old or younger. There were 40 cases = screened=20 in time with screening results missing (22 with closed = reduction,=20 two with open reduction, and 16 with osteotomy). The type of = operative procedure did not differ between children screened = within=20 the recommended timeframe or late ( 3Dchi=20

² test, p=3D0=B725). With respect to = the 495 cases=20 with information on ultrasound, 272 (55%) had abnormal = findings at=20 screening within the recommended timeframe, 64 (13%) had = normal=20 findings at screening within the recommended timeframe, 69 = (14%)=20 were screened too late, and 89 (18%) were identified outside = the=20 routine screening programme. In analysis of only cases = screened=20 within the recommended timeframe, 64 (19%) had a normal = ultrasound=20 despite being a case.=20

	Correct timing of ultrasound=20 screening=86	Delayed ultrasound screening=87	No ultrasound screening	Total
	Positive screen	Negative screen
Closed reduction	176 (53%)	44 (13%)	53 (16%)	58 (18%)	331 (100%)
Open reduction	36 (61%)	7 (12%)	6 (10%)	10 (17%)	59 (100%)
Osteotomy	60 (57%)	13 (12%)	11 (11%)	21 (20%)	105 (100%)
Total	272 (55%)	64 (13%)	70 (14%)	89 (18%)	495 (100%)
Data are number (%). *Only cases with = complete=20 information about screening results were included. = =86Up to the=20 age of 6 weeks. =87Beyond the age of 6 weeks. =
Table 2: Children with first = operative=20 procedures by time of screening and screening=20 results*

During the 5 years of observation, the incidence = calculated=20 by division of the number of recorded cases by the number of = livebirths per year was 0=B710-0=B718 per 1000 livebirths, = with a yearly=20 average of 0=B714 (95% CI 0=B713-0=B715). The = capture-recapture estimates=20 for cases born in southern Bavaria between May, 1997 and = April,=20 2001, yielded 18 cases reported only in the active = surveillance=20 programme, 13 in both data sources, and 12 were identified = only=20 during hospital visits. The capture-recapture estimate for = the true=20 number of cases was 60, and 31 of these (52%) had been = ascertained=20 by the active surveillance system. Thus, the true rate of = first=20 operative procedures is likely to be 0=B726 per 1000 (95% CI = 0=B725-0=B728). Since 272 (55%) of the cases had timely = screening and=20 abnormal findings (table 2), the true rate of first = operative=20 procedures despite timely screening and early diagnosis of=20 developmental hip dysplasia was 0=B714 cases per 1000 = livebirths.

Discussion =

The adjusted rate of first operative procedures that = we=20 calculated--0=B726 per 1000 livebirths--was considerably = lower than=20 that in other investigations,^5,9,10 and much = lower than=20 previously estimated in Germany. In 1983, before ultrasound=20 screening, Katthagen and colleagues⁸ reported 616 = first=20 operative procedures on children younger than 6 years = (yearly birth=20 rates in Germany 1977 to 1983 ranging between 590000 and = 620000),=20 accounting for a rate of about one per 1000 livebirths. Our = finding=20 of low incidence cannot be explained by our case definition, = which=20 excluded children younger than 10 weeks, since in 1983, 548 = (89%) of=20 the 616 cases with a first operative procedure were 4 months = or=20 older (mean age 9 months).⁸ Because the = historical=20 data⁸ originate from a cross-sectional survey = with one=20 mailed questionnaire only, and a participation rate of 61%, = this=20 rate is probably an underestimate.=20

Follow-up studies over more than 50 years would be = necessary=20 to assess the potential benefit of ultrasound screening for = impaired=20 hip function and premature degenerative joint disease. Thus, = previous researchers have used the rate of first operative=20 procedures during preschool age as a useful and feasible = endpoint to=20 determine the effectiveness of screening.^5,9,10 = Such an=20 assessment of the effect of clinical screening, based on = signs and=20 symptoms, led to controversial results. In one study, = investigators=20 recorded little beneficial effect of such screening (first = operative=20 procedure rate of 0=B778 per 1000 livebirths).⁵ = In another,=20 workers reported low rates of first operative procedures = (0=B746 per=20 1000 livebirths).⁹ Finally, some researchers = noted a=20 reduction in the rate of operative procedures with improved = early=20 diagnosis, from 1=B775 per 1000 livebirths to 1=B703 per = 1000=20 livebirths.¹⁰

The capture-recapture study detected considerable=20 under-reporting in our active surveillance study, and thus, = the=20 recorded rates had to be adjusted. Data for the = capture-recapture=20 analysis were obtained in hospitals by an independent = investigator=20 who was masked with respect to details of reported cases. = The=20 southern Bavarian population analysed for the = capture-recapture=20 estimate consisted of 7=B75% of all livebirths in Germany = (yearly=20 livebirth rate in southern Bavaria of 60 000 compared with = an=20 average yearly livebirth rate in Germany of 780 000) and = 7=B71% of the=20 cases observed. We believe that the data from southern = Bavaria are=20 representative of the general population because, of the 535 = cases=20 reported over 5 years across Germany, 33 would be expected = to be=20 identified by the active surveillance programme over 4 years = in this=20 area, on the basis of its annual livebirth rate. The number = recorded=20 was 31. The capture-recapture analysis showed that the = proportion of=20 cases identified by the active surveillance programme was = 51=B79%,=20 which was identical to the proportion detected in the=20 UK,⁵ and was in the range of results from other = studies=20 that used the ESPED logistics in Germany.¹¹=20

After comparison with historical data, our results = suggest=20 that an ultrasound-screening programme might substantially = reduce=20 the rate of first operation. However, historical data are = not ideal=20 for comparison, since other factors can also affect the rate = of=20 first operative procedures and might also have changed with = time.=20 Another important issue is the quality of previous clinical=20 screening in Germany to which ultrasound screening has been=20 compared. Before the introduction of ultrasound screening in = Germany, screening for developmental hip dysplasia was part = of a=20 well-baby check-up programme. Clinical screening included = the=20 Ortolani test, a check of abduction and for length = differences of=20 the leg.¹⁴ In 1982-83 a cohort of 2021 livebirths = was=20 recruited in four maternity wards in Cologne, and = followed-up for 2=20 years.¹⁵ Of these livebirths, 545 had had = radiographical=20 examination, of which 25 cases were diagnosed according to = at least=20 one radiologist's judgment. The disease had been treated in = 23;=20 however, only three were treated before the age of 4 months, = although seven had been identified by abnormal screening in = the=20 first week.=20

Results of three other investigations indicate that = an=20 ultrasound-screening programme might reduce the rate of = first=20 operative procedures. Andersson¹⁶reported in an=20 observational study a rate of 0=B715 first operative = procedures per=20 1000 livebirths for 22 047 children screened by ultrasound. = However,=20 only open reductions were considered in this study. = Investigators in=20 two randomised controlled studies^17,18 used late=20 detection of dislocation, subluxation, or acetabular = dysplasia to=20 assess the effect of screening. In the first investigation,=20 researchers compared universal ultrasound screening with = clinical=20 screening with ultrasound for at-risk children only and also = with=20 clinical screening with no ultrasound. The rates of late=20 dislocations or subluxations, were 0=B73, 0=B77, and 1=B73 = per 1000=20 livebirths, respectively.¹⁸ Researchers of = another study=20 also compared universal ultrasound screening alone with = clinical=20 screening plus ultrasound for at-risk children only. The = respective=20 rates of late detection were 0=B713 and 0=B765 per 1000=20 livebirths.¹⁷ Although universal ultrasound = screening=20 seemed to lead to improved outcome in both investigations, = the=20 differences were not substantial in either.=20

Other important results were that 19% of children who = were=20 screened within the recommended timeframe had normal = findings=20 despite being actual cases, with no evidence of geographical = clustering of those cases with a normal ultrasound. Since = review of=20 the ultrasound imaging was not possible, we could not = distinguish=20 between failures related to imaging and true screening = failures.=20 More cases were related to insufficient parental compliance = with the=20 programme, 14% because of delayed screening and 18% because = no=20 screening was done.=20

Although early treatment for developmental hip = dysplasia has=20 not been proven to prevent operative procedures, screening=20 programmes are based on the assumption that such treatment = is=20 effective. This assumption might not be true for all = children=20 identified during screening. The surveillance data collected = over 5=20 years in Germany showed that about 0=B714 (55%) of the = 0=B726 first=20 operative procedures per 1000 livebirths were for cases with = early=20 detection by an ultrasound screening programme. These = children might=20 represent a proportion of children with the disease who = might not=20 benefit from early conservative treatment. Similar rates of = first=20 operative procedure despite early diagnosis have been=20 reported.^9,10,16,19All cases with timely = diagnosis had=20 received some kind of early splinting treatment. = Unfortunately, on=20 the basis of our data and that of other=20 investigators,^9,10,16,19 differentiation between = failure=20 to respond to adequate early conservative treatment, = compliance=20 difficulties, and inadequate treatment, is not possible. = Such=20 information could only be obtained by an audit, which is = beyond the=20 scope of this questionnaire study.=20

Although the ultrasound-screening programme seems = effective=20 in reducing the rate of operative procedures, the negative=20 consequences of such a programme must be considered. An = independent=20 assessment of such consequences showed that initially, 24% = of the=20 children screened had had at least one repeat ultrasound = examination=20 of the hip,²⁰ which fell to 19% by 2000 (internal = documents of the Central Research Institute of Ambulatory = Health=20 Care in Germany). This continuing high rate of repeat = examinations=20 will be an issue in cost-effectiveness calculations. = Improved=20 training of doctors might further reduce repeat = examinations.=20 Initially, 7% of children had a recommendation for=20 treatment.²⁰ By 2000, this percentage was reduced = to 5%.=20 High indications for treatment in relation to ultrasound = screening=20 (3=B71%)have been reported,²¹ whereas rates for = defined=20 treatment (Pavlik Harness) have been noted to be as low as = 0=B724=20 %.²² Unfortunately, there are no data for = conservative=20 treatment rates for developmental hip dysplasia (eg, = splinting) in=20 Germany. In a Swedish municipality, clinical screening = resulted in=20 treatment rates of 0=B76-3=B75%.²³ If all = recommendations=20 for treatment were followed by splinting or other invasive=20 treatments, the rate of overtreatment would be unacceptably=20 high.²⁴ This issue of potential overtreatment and = side-effects of treatment should be addressed.=20

Thus, although the German ultrasound screening = programme=20 seems effective in reducing the rate of operative = procedures,=20 further improvements are needed with respect to timely = participation=20 and doctors assessment of ultrasound imaging. Better quality = data=20 for conservative treatment rates for developmental hip = dysplasia are=20 needed to monitor and possibly reduce over-treatment.

Contributors=20

R von Kries, L Altenhofen, and F Uwe Niethard were=20 responsible for concept and study design. N Ihme and A = Lorani=20 assessed the questionnaires. R Stark and D Oberle did the=20 capture-recapture study. D Oberle contributed to data = management,=20 and, with R von Kries, analysed the data. R von Kries and R = Stark=20 wrote the report.

Conflict of = interest=20 statement

None declared.

Acknowledgments=20

The authors thank all the orthopaedic surgeons who=20 contributed to this surveillance by monthly replies to our = report=20 cards and filling in the questionnaires, especially = Bernd-Dietrich=20 Katthagen (Dortmund), Klausdieter Parsch (Stuttgart), and = Franz L=F6er=20 (Essen). The help of Susanne Stoll and Klaus-Peter = G=FCnther, Dresden=20 in initiating the programme, and of Andrea Schilling in = keeping the=20 surveillance programme going was invaluable.The study was = funded by=20 Kassen=E4rztliche Bundesvereinigung (National Association of = Statutory=20 Health Insurance Physicians) and Spitzenverb=E4nde der = gesetzlichen=20 Krankenkassen (National Associations of Health Insurance = Funds,=20 Sickness Funds).

References =

1 Pauwels F. Biomechanics of the normal and diseased = hip.=20 Berlin: Springer, 1976.=20

2 Barlow TG. Early diagnosis and treatment of = congenital=20 dislocation of the hip joint. J Bone Jt=20 Surg 1962; 44: 292-301. [PubMed]=20

3 von Rosen S. Early diagnosis and treatment of = congenital=20 dislocation of the hip joint. Acta Orthop=20 Scand 1956; 26: 136-40. [PubMed]=20

4 Ortolani M. Un segno poco noto es sua importanza = per la=20 dignosi precoce de prelussazione congenita dell'anca.=20 = Pediatri 1937; 45: 129-34. [PubMed]=20

5 Godward S, Dezateux C. Surgery for congenital = dislocation=20 of the hip in the UK as a measure of outcome of screening.=20 = Lancet 1998; 351: 1149-52. [Text]=20

6 Graf R. New possibilities for the diagnosis of = congenital=20 hip joint dislocation by ultrasonography. J Pediatr = Orthop 1983; 3: 354-59. [PubMed]=20

7 Altenhofen LH. Leitlinie f=FCr das = h=FCftsonographische=20 Screening im Rahmen des Programms Krankheitsfr=FCherkennung = im=20 Kindesalter, Leitlinie Konsensusstatement. = Deutsches=20 =C4rzteblatt = 1996; 93: 57-60. [PubMed]=20

8 Katthagen BD, Mittermeier H, Becker D. H=E4ufigkeit = und=20 station=E4rer Behandlungsbeginn kindlicher = H=FCftgelenksluxationen in=20 der BR Deuschland. Z=20 Orthop 1988; 126: 475- = 83. [PubMed]=20

9 Chan A, Cundy PJ, Foster BK, Keane RJ, Byron-Scott = R. Late=20 diagnosis of congenital dislocation of the hip and presence = of a=20 screening programme: South Australian population-based = study.=20 = Lancet 1999; 354: 1514-17. [Text]=20

10 Maxwell SL, Ruiz AL, Lappin KJ, Cosgrove AP. = Clinical=20 screening for developmental dysplasia of the hip in Northern = Ireland.=20 = BMJ 2002; 324: 1031-33. [PubMed]=20

11 von Kries R, Heinrich B, Hermann M. P=E4diatrische = Epidemiologie in Deutschland: Forschungsinstrument ESPED=20 (Erhebungseinheit f=FCr seltene p=E4diatrische Erkrankungen = in=20 Deutschland). Monatsschr=20 = Kinderheilkd 2001; 149: 1191-97. [PubMed]=20

12 Hook EB, Regal RR. Capture-Recapture methods in=20 epidemiology: Methods and limitations. Epidemiol=20 Rev 1995; 17: 243-64. [PubMed]=20

13 Laubereau B, Hermann M, Weil J, Schmitt HJ, von = Kries R.=20 Durchimpfungsraten bei Kindern in Deutschland 1999: = Grunds=E4tzliche=20 Impfbereitschaft, aber Impfungen h=E4ufig zu sp=E4t und = inkomplett.=20 Monatsschr Kinderheilkd=20 2001; 149: 367-73. [PubMed]=20

14 Zentralinstitut. Hinweise zur Durchf=FChrung der=20 Fr=FCherkennungsuntersuchungen im Kindesalter. K=F6ln: = Deutscher=20 =C4rzte-Verlag, 1991.=20

15 Weidtman V, von Heeremann B, Herrmann F. = Ermittlung der=20 Ursachen von Fehlbeurteilungen angeborener = H=FCftgelenksver=E4nderungen=20 zur Operationalisierung optimaler Screeningstrategien. = K=F6ln:=20 Bundesministerium f=FCr Forschung und Technologie, 1984.=20

16 Andersson JE. Neonatal hip instability: results = and=20 experiences from ten years of screening with the = anterior-dynamic=20 ultrasound method. Acta Paediatr=20 2002; 91: 926-29. [PubMed]=20

17 Holen KJ, Tegnander A, Bredland T, et al. = Universal or=20 selective screening of the neonatal hip using ultrasound? = J=20 Bone Joint Surg=20 2002; 84: 886-90. [PubMed]=20

18 Rosendahl K. MT, Lie R.T. Ultrasound screening for = developmental dysplasia of the hip in the neonate: the = effect on=20 treatment rate and prevalence of late cases.=20 = Pediatrics 1994; 94: 47-52. [PubMed]=20

19 Elbourne D, Desateux C, Arthur R, et al. = Ultrasonography=20 in the diagnosis and management of developmental hip = dysplasia (UK=20 Hip Trial): clinical and economical results of a multicenter = randomised controlled trial.=20 = Lancet 2002; 360: 2009-17. [Text]=20

20 Niethard FU, G=FCnther KP, von Kries R, Altenhofen = L.=20 Klinisches und sonographisches Screening der = S=E4uglingsh=FCfte.=20 Dt =C4rztebl=20 2000; 97: 1593-99. [PubMed]=20

21 Rosendahl K, Markestad T, Lie RT. Congenital = dislocation=20 of the hip: a prospective study comparing ultrasound and = clinical=20 examination. Acta Paediatr=20 1992; 81: 177-81. [PubMed]=20

22 Marks DS, Clegg J, al-Chalabi AN. Routine = ultrasound=20 screening for neonatal hip instability. Can it abolish=20 late-presenting congenital dislocation of the hip? J = Bone=20 Joint Surg = Br 1994; 76: 534-38. [PubMed]=20

23 Duppe H, Danielsson LG. Screening of neonatal = instability=20 and of developmental dislocation of the hip: a survey of = 132,601=20 living newborn infants between 1956 and 1999. J Bone = Joint=20 Surg Br 2002; 84: 878-85. [PubMed]=20

24 Eastwood DM. Neonatal hipscreening.=20 = Lancet 2003; 361: 595-97. [Text]

= =A9 2003 Elsevier Limited. THE = LANCET=AE is a=20 registered trademark of Elsevier Properties S.A.	= \| Privacy	= \| Terms=20 & Conditions