Evaluation of Brace Treatment for Infant Hip Dislocation

Evaluation of Brace Treatment for Infant Hip Dislocation in a Prospective Cohort

Defining the Success Rate and Variables Associated with Failure

Vidyadhar V. Upasani, MD, James D. Bomar, MPH, Travis H. Matheney, MD, Wudbhav N. Sankar, MD, Kishore Mulpuri, MBBS, MS(Ortho), MHSc(Epi), Charles T. Price, MD, Colin F. Moseley, MD, CM, FRCS, Simon P. Kelley, MBChB, MRCS, FRCS, Unni Narayanan, MBBS, MSc, FRCS(C), Nicholas M.P. Clarke, ChM, DM, FRCS, John H. Wedge, OC, MD, FRCSC, Pablo Castañeda, MD, James R. Kasser, MD, Bruce K. Foster, MBBS, MD, FRACS, Jose A. Herrera-Soto, MD, Peter J. Cundy, MBBS, FRACS, Nicole Williams, FRACS, BMed, BMedSc, and Scott J. Mubarak, MD

Background: The use of a brace has been shown to be an effective treatment for hip dislocation in infants; however, previous studies of such treatment have been single-center or retrospective. The purpose of the current study was to evaluate the success rate for brace use in the treatment of infant hip dislocation in an international, multicenter, prospective cohort, and to identify the variables associated with brace failure.

Methods: All dislocations were verified with use of ultrasound or radiography prior to the initiation of treatment, and patients were followed prospectively for a minimum of 18 months. Successful treatment was defined as the use of a brace that resulted in a clinically and radiographically reduced hip, without surgical intervention. The Mann-Whitney test, chisquare analysis, and Fisher exact test were used to identify risk factors for brace failure. A multivariate logistic regression model was used to determine the probability of brace failure according to the risk factors identified.

Results: Brace treatment was successful in 162 (79%) of the 204 dislocated hips in this series. Six variables were found to be significant risk factors for failure: developing femoral nerve palsy during brace treatment (p = 0.001), treatment with a static brace (p < 0.001), an initially irreducible hip (p < 0.001), treatment initiated after the age of 7 weeks (p = 0.005), a right hip dislocation (p = 0.006), and a Graf-IV hip (p = 0.02). Hips with no risk factors had a 3% probability of failure, whereas hips with 4 or 5 risk factors had a 100% probability of failure.

Conclusions: These data provide valuable information for patient families and their providers regarding the important variables that influence successful brace treatment for dislocated hips in infants.

Level of Evidence: Prognostic Level I. See Instructions for Authors for a complete description of levels of evidence.

Developmental dysplasia of the hip (DDH) is the most common hip disorder in the pediatric population, with 1% to 3% of all newborns being diagnosed at birth1,2.This diagnosis describes a wide spectrum of abnormal hip morphology ranging from mild (a reduced femoral head in a dysplastic ace-tabulum) to severe (a femoral head dislocated entirely from the acetabulum). The dislocated femoral head might be passively reducible (Ortolani positive) or irreducible (Ortolani negative) as a result of muscle contractures and intra-articular obstructions3-5.

The natural history of an untreated unilateral hip dislocation has been studied6,7. Most such patients have a poor clinical outcome due to limb-length inequality, ipsilateral knee deformity and pain, or secondary scoliosis and back pain. However, most patients with DDH can be effectively treated if diagnosed at a young age, justifying some form of neonatal screening. While universal neonatal ultrasonography is performed in some European countries, other countries, including the United States, perform selective screening of infants with risk factors for hip dysplasia (e.g., first-born status, female sex, breech intrauterine position, or a family history of DDH) or when findings on physical examination are consistent with hip instability8-12.

The primary goals in the treatment of a dislocated infant hip are well established: (1) to obtain concentric reduction of the joint, (2) to maintain reduction, and (3) to allow for appropriate development of the femoral head and acetabulum, while (4) avoiding complications such as osteonecrosis of the femoral head. Treatment with a brace previously has been shown to be a variably effective strategy to successfully manage dislocated hips13-20. Previous studies on the Pavlik harness, the mostly widely used orthosis for DDH in infants, have demonstrated success rates ranging from 58% to 97%17-20. Most of these studies, however, have been retrospective, from a single institution, and without clear documentation of the severity of hip dislocation prior to initiating brace treatment18-22.

The goals of the present study were to evaluate the success rate of brace treatment for infants with evidence of hip dislo-cation in an international, multicenter, prospective cohort, and to identify the variables associated with brace failure. We hypothesized that brace treatment could successfully manage the majority of the dislocated hips and that specific risk factors could be identified to determine the probability of brace failure.

Peer review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. It was also reviewed by an expert in methodology and statistics. The Deputy Editor reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or more exchanges between the author(s) and copyeditors.

Disclosure: REDCap database coordination, maintenance, and support was provided by the International Hip Dysplasia Institute. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work and “yes” to indicate that the author had other relationships or activities that could be perceived to influence, or have the potential to influence, what was written in this work.

Materials and Methods

Study Design

A research network for the prospective study of DDH was established involving 7 institutions across North America, Europe, and Australia. Institutional review board or ethics committee approval was granted at all sites prior to the initiation of data collection. Study data were collected prospectively and managed using REDCap (Research Electronic Data Capture) tools23. The group standardized the diagnostic criteria used to identify the dislocated hips; however, each center was allowed to use independent treatment protocols as this would permit the study of practice variation and outcomes.

Inclusion Criteria

We enrolled all infants who were <6 months of age and had a new diagnosis of hip dislocation at rest. The hip dislocation was verified with use of ultrasound or radiography prior to the initiation of treatment. Hips were considered dis-located if the femoral head was <30% covered on the coronal view on ultra-sound or if they were IHDI (International Hip Dysplasia Institute) grade III or IV on radiographs24. Patients were followed for at least 18 months, by which time they had either completed treatment with no evidence of requiring further treatment or were treated surgically after failing brace treatment. Nine hips with <18 months of follow-up were included because they required surgery prior to 18 months, which was considered a terminal event.

Exclusion Criteria

Infants were excluded if they were enrolled in the study but their family refused treatment; if the dislocation was associated with a syndrome or other congenital hip abnormality; if they had a milder form of DDH, such as a subluxable or dysplastic hip with no dislocation; or if they had received previous treatment for DDH. Additionally, 13 hips were excluded from this analysis because they were initially treated surgically with no attempt at brace treatment (Fig. 1); 8 of the hips were treated with closed reduction, and 5 were treated with open reduction. Patients were considered lost to follow-up if they were enrolled at least 18 months prior to the initiation of this analysis but did not have 18 months of follow-up and were not seen in the 6 months prior to this analysis. All of the centers tried various strategies to contact the families of patients lost to follow-up, including telephone calls, letters, and electronic messages.

Patient Cohort

Two hundred and four dislocated hips in 159 infants were evaluated.

Data Collection

Demographic details, such as date of birth, maternal age, maternal parity, birth weight, sex, gestational age, fetal presentation, birth method, family history, and swaddling history, were recorded at baseline. The results of a clinical examination, including an assessment of hip abduction and hip reducibility and stability, and motor and sensory examination, were recorded at each visit. Ultrasonographic data collected included the location of the femoral head, joint laxity, percentage of coverage of the femoral head, and assessment of osseous acetabular morphology and the cartilage roof triangle. The alpha angle was evaluated as a categorical variable according to our osseous acetabular grades (A = good, alpha angle of >59;B= deficient/rounded, alpha angle of 50 to 59;and C= dysplastic, alpha angle of <50). Each hip was also graded according to the Graf classification25; the dislocated hips in this cohort were given a Graf grade of III or IV on the basis of the shape of the osseous acetabulum and the cartilage roof triangle on a static coronal image. If the cartilage roof triangle was interposed, the hip was classified as Graf grade IV. Radiographic data collected included evaluation of the Shenton line, the acetabular index, the IHDI grade24, the state of ossification of the femoral head, a description of the teardrop, the lateral center-edge angle, and the presence of osteonecrosis of the femoral head.

Treatment details, including the age at the initiation of treatment, the duration of treatment, the type of brace and the number of hours that brace wear was prescribed, and any surgical intervention or complications, were recorded. Static braces, such as the Denis Browne, Von Rosen, and Plastazote braces, were grouped together and compared with treatment with the dynamic Pavlik harness.

Primary Outcome

Successful treatment was defined as obtaining and maintaining reduction of the hip following bracing with no subsequent surgical treatment. Hip reduction was determined radiographically using the IHDI grading system24 on the most recent anteroposterior pelvic radiograph, and clinically based on a stable and unrestricted range of hip abduction on examination. Failure was defined as a hip that did not achieve or maintain reduction of the hip joint at the time of final follow-up after being treated with bracing alone.

Secondary Outcomes

Secondary outcomes included the acetabular index as a measure of residual acetabular dysplasia, and complications such as femoral nerve palsy (clinical diagnosis) and osteonecrosis (radiographic).

Statistical Analysis

Significant differences were noted between right and left hips in bilateral cases. Forty percent of the bilateral cases differed between right and left with respect to reducibility and osseous or cartilaginous classifications, and therefore, the hip was used as the unit of analysis to determine variables associated with our primary outcome. The Shapiro-Wilk test of normality was performed on all continuous data. The Mann-Whitney test was used to compare differences in age at the initiation of treatment between hips that failed brace treatment and those that did not. Chi-square analysis and the Fisher exact test were used to evaluate brace failures on the basis of the reducibility of the hip, the severity of the dislocation (Graf grade and alpha angle), bilaterality, brace type (static versus dynamic), the rate of complica-tions, and treatment facility (by individual site and as grouped according to whether the site had in-clinic ultrasound and according to volume). Variables associated with our primary outcome at a significance level of p < 0.1 were included in a multi-variate logistic regression model to evaluate predictors of brace treatment failure. Adjusted odds ratios (ORs) derived from the logistic regression model were re-ported. A multivariate generalized estimating equation (GEE) was used to control for the potential effect of bilateral cases on the model. A receiver operating char-acteristic (ROC) curve was constructed to evaluate the accuracy of the risk factors in our model in their ability to predict brace failure. An analysis of variance (ANOVA) was used to compare differences in the acetabular index at the time of final follow-up between hips that failed brace treatment and those that did not. No a priori power analysis was performed. All statistics were calculated using SPSS software (version 22; IBM), with significance defined as p < 0.05.

Results

Demographics are presented in Table I. The mean age at the initiation of brace treatment (and standard deviation) was 39 ± 36 days (range, 0 to 163 days). The mean duration of follow-up was 27 ± 8 months (range, 6 to 49 months). One hundred and sixty-two of the 204 hips were successfully treated with a brace (a 79% success rate). Forty-two hips failed brace treatment and required surgical treatment (a 21% failure rate). The mean age at the initiation of treatment for hips that were successfully treated with a brace was 35 ± 34 days compared with 54 ± 40 days for the hips that failed brace treatment (p = 0.006). The average acetabular index at the time of final follow-up was 22 ± 4 (range, 11 to 31) among the hips successfully treated with a brace and 26 ± 5 (range, 13 to 35) among the hips that required surgical treatment (p < 0.001). Ten (5%) of the 204 hips in this cohort were noted to have radiographic evidence of osteonecrosis of the femoral head.

The results of brace treatment by site are presented in Figure 2. Success rates ranged from 50% to 88% across the 7 centers. One facility (G) had a significantly higher success rate (88%) compared with that of the other sites combined (74%)(p = 0.02). When treatment facilities were grouped by those that had in-clinic ultrasound monitoring (C, D, and G) and those that did not (A, B, E, and F), no significant difference in brace success was observed (Table II) (p = 0.08). Brace out-comes were not found to be significantly different at the three highest-volume sites (E, F, and G) compared with the others (A, B, C, and D) (p = 0.119).

In the univariate analysis, 8 variables met the significance criteria to be included in the multivariate model (Table II). The proportion of hips that failed treatment did not differ on the basis of whether the patient had bilateral or unilateral in-volvement (p = 0.9) or on the basis of sex (p = 1.0). Six vari-ables were found to be significant risk factors for failure in the multivariate model (Table III). A test of the full model dem-onstrated significance, indicating that the predictors as a set reliably distinguished between success and failure of brace treatment (chi square = 72.7; degrees of freedom = 8; p < 0.001). The model accounted for 30% to 48% of variation. The sensitivity of the model in predicting failure was 94%, and the specificity was 56%. The probability of brace failure was as-sessed according to the number of risk factors present (Table III). Three percent of the hips with no risk factors went on to fail brace treatment compared with 100% of hips with 4 or 5 risk factors. In constructing an ROC curve, the area under the curve was found to be 0.83 (p < 0.001).

Loss to Follow-up

Forty-three infants with 54 dislocated hips (21%) were lost to follow-up during treatment and were excluded from the analysis (Table I). The mean follow-up for this cohort was 9 ± 4 months (range, 0.9 to 16 months). None of these patients were found to have osteonecrosis or femoral nerve palsy prior to being lost to follow-up.

Discussion

This is the first study that we know of to assess the success of brace treatment for hip dislocation in infants using an international, multicenter, prospective cohort study design. Unlike previous investigations, this study included a clearly defined prospective cohort of infants with only dislocated hips that were either reducible (Ortolani positive) or irreducible (Ortolani negative). All patients had radiographic or ultraso-nographic verification of the hip dislocation prior to the ini-tiation of treatment, and patients were followed for at least 18 months. The overall success rate of treatment with a brace for this cohort was 79%.

Reducibility of the hip has consistently been shown to be an important variable in brace success18-22. In the current cohort, ultrasonography was used to verify whether the femoral head was reducible or irreducible with the Ortolani maneuver (gentle hip traction, flexion, and abduction) at initial presentation. While a previous study recommended abandoning harness treatment for these irreducible dislocations26, we demonstrated a 55% early success rate among irreducible hips treated with use of a brace. It is important to follow these complicated patients closely with weekly ultrasound and clinical examinations to ensure that the range of hip abduction is improving and that femoral nerve function is intact, and to assess the position of the femoral head. Longer follow-up is needed to better understand rates of osteo-necrosis in these hips.

Patient age at the initiation of brace treatment was eval-uated as a continuous variable and as a categorical variable. The mean age at the initiation of treatment among the hips that were successfully treated with a brace was significantly less than that among the hips that failed brace treatment (p = 0.006). Atalar et al. evaluated 31 frankly dislocated hips in 25 patients documented by dynamic ultrasonography21.They found a significantly higher success rate with bracing if treatment was initiated prior to the age of 7 weeks (p = 0.038). On the basis of their data, we also used a 7-week cutoff to evaluate our patient cohort and found a significant difference between the two groups. In our cohort, hips treated after the age of 49 days were 4 times more likely to fail brace treatment. This finding dem-onstrates the importance of appropriate education of pedia-tricians and neonatologists in conducting routine, serial, perinatal hip examinations and of the urgent referral of dis-located hips to optimize brace treatment outcomes. It is important to emphasize that these data support early treatment for only the most severe dislocated infant hips, which represent a minority (1% to 5%) of all patients with DDH27.

It is unclear why the side of the hip dislocation was a significant risk factor in the multivariate analysis. As in pre-vious investigations, a majority (56%) of the dislocated hips in this cohort were left-sided. It is hypothesized that left-sided dislocations are more common because of the left occiput anterior fetal positioning, which causes increased adduction of the infant’s left hip in utero against the maternal sacrum28.It could be that right-sided hip dislocations are inherently more severe and thus more difficult to treat conservatively with a brace. Further analysis of this finding is required to better understand its importance.

The vast majority (93%) of the hips in this cohort were treated with a Pavlik harness, with an overall success rate of 83%. Conversely, 9 of 14 hips initially treated with a static brace went on to require surgical treatment (a 36% success rate). The brace types were grouped together in the statistical analysis because of the similar mechanism of correction (simple ab-duction with the static braces versus dynamic flexion and abduction with the Pavlik harness)14,29-32. The indication for using a static abduction brace varied among centers, ranging from standard practice to family preference. Although the discrep-ancy in the sample size of the two groups was substantial, the statistical methods used allowed for appropriate comparison between the two treatment methods.

All patients in this cohort who developed femoral nerve palsy (8 hips) were treated with a Pavlik harness, and the nerve palsy was likely due to excessive flexion of the hip. The 4% rate of this complication is slightly greater than the previously reported rate of 2.7%33. This may be because all patients in our cohort had more severe DDH, with completely dislocated hips, while the previous study included milder forms of dysplasia33. Of the 8 hips with femoral nerve palsy, 6 (75%) failed brace treatment and underwent surgery. This may indicate a selection bias, as brace treatment was discontinued for these patients for a variable period of time to allow for recovery of nerve function. However, it may also indicate that these hips were in-herently more difficult to reduce or keep reduced, requiring hyperflexion of the hip in the brace, which may have caused the femoral nerve palsy. Patient age at the initiation of treatment did not seem to play a role; among the 8 hips for which femoral nerve palsy was noted, treatment was initiated at an average of 60 days (range, 12 to 81 days; median, 66 days) for the 6 hips that failed treatment and at 12 days and 71 days for the 2 hips that were successfully treated with a brace.

There were limitations to this study that should be ad-dressed. Primarily, 21% of the patients were excluded because they were lost to follow-up, despite our best attempts to contact the families. At the latest follow-up, however, a majority of the hips were reduced and the lost-to-follow-up group appeared to be comparable with the study cohort in terms of the success rate of brace treatment. Also, the risk factors identified in this study were good predictors of failure in this cohort because we built the model around this data set; additional studies, involving a different cohort, should be carried out to determine the generalizability of these risk factors as predictors of failure. Additionally, although the average follow-up for this cohort was more than 2 years, the ultimate success of brace treatment needs to be followed until skeletal maturity to determine the true rate of femoral head osteonecrosis and the ultimate need for surgical treatment. Another limitation was that there was no central adjudication of radiographic and ultrasonographic measurements in this study group.

In this prospective, international, multicenter cohort study, we demonstrated a higher than previously reported success rate of the Pavlik harness as the initial management of a dislocated infant hip (not associated with a syndrome or congenital malformation other than DDH). The multivariate analysis identified specific risk factors associated with higher failure rates, which, for the first time that we know of, provide treating physicians explicit prognostic data with respect to the success of nonoperative treatment and the likelihood of requiring subsequent surgical management. Additional investigation is necessary to elucidate finer details of the duration of brace wear and when to abandon brace treatment when faced with persistent dislocation, in order to maximize success of nonoperative management and minimize complications.

NOTE: The authors thank Emily Schaeffer, PhD, for data management and analysis.

References
1. Godward S, Dezateux C; MRC Working Party on Congenital Dislocation of the Hip. Medical Research Council. Surgery for congenital dislocation of the hip in the UK as a measure of outcome of screening. Lancet. 1998 Apr 18;351
(9110):1149-52.
2. Tibrewal S, Gulati V, Ramachandran M. The Pavlik method: a systematic review of current concepts. J Pediatr Orthop B. 2013 Nov;22(6):516-20.
3. Ortolani M. [A very little known sign and its importance in the early diagnosis of congenital hip predislocation]. Atti Accad Medica Ferra. 1936. Italian.
4. Ortolani M. [A little known sign and its importance for the diagnosis early sub-luxation in congenital hip]. La Pediatr. 1937;129. Italian.
5. Ortolani M. Congenital hip dysplasia in the light of early and very early diagnosis. Clin Orthop Relat Res. 1976 Sep;119:6-10.
6. Weinstein SL. Natural history of congenital hip dislocation (CDH) and hip dys-plasia. Clin Orthop Relat Res. 1987 Dec;225:62-76.
7. Wedge JH, Wasylenko MJ. The natural history of congenital dislocation of the hip: a critical review. Clin Orthop Relat Res. 1978 Nov-Dec;137:154-62.
8. Clarke NMP, Reading IC, Corbin C, Taylor CC, Bochmann T. Twenty years expe-rience of selective secondary ultrasound screening for congenital dislocation of the hip. Arch Dis Child. 2012 May;97(5):423-9. Epub 2012 Mar 12.
9. US Preventive Services Task Force. Screening for developmental dysplasia of the hip: recommendation statement. Pediatrics. 2006 Mar;117(3):898-902.
10. U.S. Preventive Service Task Force. Screening for developmental dysplasia of the hip: recommendation statement. Am Fam Physician. 2006 Jun 1;73
(11):1992-6.
11. Shipman SA, Helfand M, Moyer VA, Yawn BP. Screening for developmental dysplasia of the hip: a systematic literature review for the US Preventive Services Task Force. Pediatrics. 2006 Mar;117(3):e557-76.
12. Rosendahl K, Markestad T, Lie RT. Developmental dysplasia of the hip. A population-based comparison of ultrasound and clinical findings. Acta Paediatr. 1996 Jan;85(1):64-9.
13. Cashman JP, Round J, Taylor G, Clarke NMP. The natural history of devel-opmental dysplasia of the hip after early supervised treatment in the Pavlik harness. A prospective, longitudinal follow-up. J Bone Joint Surg Br. 2002 Apr;84 (3):418-25.
14. Pavlik A. Stirrups as an aid in the treatment of congenital dysplasias of the hip in children. By Arnold Pavlik, 1950. J Pediatr Orthop. 1989 Mar-Apr;9(2):157-9.
15. von Rosen S. Diagnosis and treatment of congenital dislocation of the hip hoint in the new-born. J Bone Joint Surg Br. 1962 May;44(2):284-91.
16. Browne D. The treatment of congenital dislocation of the hip. Proc R Soc Med. 1948 Jun;41(6):388-90.
17. White KK, Sucato DJ, Agrawal S, Browne R. Ultrasonographic findings in hips with a positive Ortolani sign and their relationship to Pavlik harness failure. J Bone Joint Surg Am. 2010 Jan;92(1):113-20.
18. Lerman JA, Emans JB, Millis MB, Share J, Zurakowski D, Kasser JR. Early failure of Pavlik harness treatment for developmental hip dysplasia: clinical and ultrasound predictors. J Pediatr Orthop. 2001 May-Jun;21(3):348-53.
19. Viere RG, Birch JG, Herring JA, Roach JW, Johnston CE. Use of the Pavlik har-ness in congenital dislocation of the hip. An analysis of failures of treatment. J Bone Joint Surg Am. 1990 Feb;72(2):238-44.
20. Swaroop VT, Mubarak SJ. Difficult-to-treat Ortolani-positive hip: improved suc-cess with new treatment protocol. J Pediatr Orthop. 2009 Apr-May;29(3):224-30.
21. Atalar H, Sayli U, Yavuz OY, Uras¸ I, Dogruel H. Indicators of successful use of the Pavlik harness in infants with developmental dysplasia of the hip. Int Orthop. 2007 Apr;31(2):145-50. Epub 2006 Apr 7.
22. Palocaren T, Rogers K, Haumont T, Grissom L, Thacker MM. High failure rate of the Pavlik harness in dislocated hips: is it bilaterality? J Pediatr Orthop. 2013 Jul-Aug;33(5):530-5.
23. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research elec-tronic data capture (REDCap)—a metadata-driven methodology and workflow pro-cess for providing translational research informatics support. J Biomed Inform. 2009 Apr;42(2):377-81. Epub 2008 Sep 30.
24. Narayanan U, Mulpuri K, Sankar WN, Clarke NM, Hosalkar H, Price CT; Inter-national Hip Dysplasia Institute. Reliability of a new radiographic classification for developmental dysplasia of the hip. J Pediatr Orthop. 2015 Jul-Aug;35(5):478-84.
25. Graf R. The diagnosis of congenital hip-joint dislocation by the ultrasonic Combound treatment. Arch Orthop Trauma Surg. 1980;97(2):117-33.
26. Mostert AK, Tulp NJ, Castelein RM. Results of Pavlik harness treatment for neonatal hip dislocation as related to Graf’s sonographic classification. J Pediatr Orthop. 2000 May-Jun;20(3):306-10.
27. Weinstein SL. Developmental hip dysplasia and dislocation. In: Weinstein SL, Flynn JM, eds. Lovell and Winter’s pediatric orthopaedics. 7th ed. Philadelphia: Lippincott Williams & Wilkins; 2014:983-1111.
28. Dunn PM. Perinatal observations on the etiology of congenital dislocation of the hip. Clin Orthop Relat Res. 1976 Sep;119:11-22.
29. Harris IE, Dickens R, Menelaus MB. Use of the Pavlik harness for hip displace-ments. When to abandon treatment. Clin Orthop Relat Res. 1992 Aug;281:29-33.
30. McKibbin B. Anatomical factors in the stability of the hip joint in the newborn. J Bone Joint Surg Br. 1970 Feb;52(1):148-59.
31. Nakamura J, Kamegaya M, Saisu T, Someya M, Koizumi W, Moriya H. Treatment for developmental dysplasia of the hip using the Pavlik harness: long-term results. J Bone Joint Surg Br. 2007 Feb;89(2):230-5.
32. Filipe G, Carlioz H. Use of the Pavlik harness in treating congenital dislocation of the hip. J Pediatr Orthop. 1982 Oct;2(4):357-62.
33. Murnaghan ML, Browne RH, Sucato DJ, Birch J. Femoral nerve palsy in Pavlik harness treatment for developmental dysplasia of the hip. J Bone Joint Surg Am. 2011 Mar 2;93(5):493.