Key points

Introduction

Developmental dysplasia of the hip (DDH) is a spectrum of hip disorders that range from a mildly dysplastic but concentrically reduced and stable hip, to a hip that is frankly dislocated and severely dysplastic ( Fig. 1 ). The goal in the management of DDH is to achieve a stable, concentric reduction of the hip to ensure that any dysplasia is adequately corrected and to avoid the complications of treatment, the most significant of which is avascular necrosis (AVN) of the femoral head.

AP radiograph of 4-year-old pelvis showing bilateral dislocated hips.

The treatment options for DDH vary depending on the age at presentation and where along the spectrum of disease the patient’s condition lies. This article reviews the evidence for the treatment of DDH. The major outcome measures assessed are the failure of reduction or redislocation, additional surgery, and AVN.

Introduction

Developmental dysplasia of the hip (DDH) is a spectrum of hip disorders that range from a mildly dysplastic but concentrically reduced and stable hip, to a hip that is frankly dislocated and severely dysplastic ( Fig. 1 ). The goal in the management of DDH is to achieve a stable, concentric reduction of the hip to ensure that any dysplasia is adequately corrected and to avoid the complications of treatment, the most significant of which is avascular necrosis (AVN) of the femoral head.

AP radiograph of 4-year-old pelvis showing bilateral dislocated hips.

The treatment options for DDH vary depending on the age at presentation and where along the spectrum of disease the patient’s condition lies. This article reviews the evidence for the treatment of DDH. The major outcome measures assessed are the failure of reduction or redislocation, additional surgery, and AVN.

Pavlik harness

Arnold Pavlik first reported the use of the Pavlik harness in 1946 and revolutionized the treatment of hip dysplasia. His theory was that hip abduction and knee flexion would maintain hip motion and result in the gentle reduction of the hip. Successful treatment, defined as reduction of a dislocated hip or maintenance of a reduced but dysplastic hip, has been reported at rates of 80.2% to 100% ( Table 1 ). It is stated that treatment should be started as soon as the diagnosis of hip dysplasia is made. The earlier the diagnosis, the higher the success rate, with statistically significant improvements seen if treatment is started before 7 weeks of age. The natural history of hip instability is one of potential spontaneous resolution in the early weeks of life and, as such, the effect size of the Pavlik harness may be overestimated. Several studies, including 1 level of evidence at level 2 and 1 at level 3, have compared immediate bracing with ultrasonographic examination and observation only, which have shown no difference in outcome if bracing is delayed for 2 weeks from birth.

The reported duration of treatment in the Pavlik harness varied from 11 to 28 weeks, with no consensus on the duration of treatment. Some investigators recommend weaning of the brace following 6 weeks of location of the hip. Reported time to reduction varied from 2 weeks to 32 weeks. Ultrasonography is an important tool for the surveillance of treatment and to determine when treatment should be discontinued either because of successful or unsuccessful treatment.

The AVN rate following treatment with the Pavlik harness has been reported at rates varying from 0% to 28%. Two studies have suggested a correlation between increasing severity of dislocation and AVN rate. Femoral nerve palsy is a known complication of Pavlik harness treatment. Murnaghan and colleagues reported an AVN rate of 2.5% of 1218 patients and a 70% chance of failure of Pavlik treatment in the presence of a femoral nerve palsy.

Residual acetabular dysplasia has been reported following Pavlik harness treatment. Harris and colleagues reported a 5% rate of acetabular dysplasia in 550 patients at 2 years. Fujioka and colleagues examined 158 hips at a minimum of 20 years’ follow-up and found that 26.6% of patients were dysplastic as defined by a Severin classification of grade III or higher, with 2.5% classified as grade V (in which the femoral head articulates with a pseudoacetabulum). Yoshitaka and colleagues found similar results with 26.3% of 262 hips rated as Severin class III or IV at a mean follow-up of 19 years. Tucci and colleagues found that 17% of 74 hips initially treated by Pavlik harness for acetabular dysplasia had what were described as mild abnormalities of the acetabular roof at an average of 10 years’ follow-up. According to the Severin classification, 1 hip was class II and 1 hip was class III. All patients had normal radiographs at ages 1 and 5 years.

The Pavlik harness has been compared with other methods of treatment. Atar and colleagues compared the Pavlik harness with the Frejka pillow in a nonrandomized case series, evaluating reduction and AVN rate. The case series found a failure rate of 12% for the Pavlik and 10% for the Frejka with corresponding AVN rates of 6% and 7%. Czubak and colleagues compared patients older than 6 months; 143 patients using the Frejka pillow and 95 using the Pavlik harness. Results showed that 89% of the hips treated with the Frejka pillow and 95% treated with the Pavlik harness were successfully reduced, with AVN rates of 7% in the Frejka group and 12% in the Pavlik group, and it was concluded that there were better results for the Frejka pillow. However, Brurås and colleagues found no difference in rates of dysplasia in a randomized control trial of Frejka pillow followed by rigid abduction bracing compared with surveillance only at 6 years.

Wilkinson and colleagues retrospectively compared the efficacy of the Pavlik harness, the Craig splint, and the von Rosen splint with no splint in the treatment of neonatal hip dysplasia. In this nonrandomized study treatment was decided on by the treating orthopedic surgeon. They concluded that the von Rosen splint resulted in better radiological appearance and fewer secondary interventions than the other forms of treatment; however, the numbers were small (the largest study arm had only 43 hips) with multiple potential confounding factors.

The von Rosen splint has been reported to have a success rate (defined as no requirement for further intervention) between 92% and 100%, with an AVN rate of 0.6% to 0.8% ( Table 2 ). These outcome measures have limitations: the study groups were heterogenous, timing of treatment varied, and the presence of AVN was not routinely recorded.

Closed reduction

The closed reduction of the dislocated hip was first described in 1838 by Pravez and often involves examination under anesthesia, an arthrogram either with or without an adductor tenotomy, followed by immobilization of the hip, usually using a spica cast. Preoperative traction can be used as an aid to reduction.

Failed reduction following closed reduction has been reported at rates varying from 0% to 25%, with redislocation rates reported between 2.8% and 13.6%. Rates of AVN have been reported as 2.6% and 60%, with the variation largely caused by differences in the definitions of AVN and the timing of follow-up evaluation. The rate of further surgical procedures is again highly variable and is based on the different treatment rationales and thresholds for surgical intervention. These rates have been reported between 10% and 101%. For example, Terjesen and Halvorsen reported 79 procedures in 78 patients, largely because of their practice of performing derotation osteotomies of the proximal femur as a staged procedure following closed reduction.

Traction has been described either as a definitive method of reducing a dislocated hip or as a preoperative adjunct to closed reduction. The evidence for preoperative traction is conflicting. Gregosiewicz and Wosko reported a rate of AVN of 6.5% with traction compared with 16% without traction. DeRosa and Feller were proponents of preoperative traction as an aid to reducing the hips of 60 of 66 patients with no AVN. Daoud and Saighi-Bououina also reported a successful closed reduction rate in older children (mean age of 33 months) of 76% of 50 hips with an AVN rate of 7.9% and a redislocation rate of 7.9%. Neither of these studies had a control group for comparison.

Langenskiöld and Paavilainen compared 176 hips treated with preoperative traction with a historical group of 86 hips treated before the introduction of traction in their hospital. The AVN rate in the patients with traction was 21.6% and without traction it was higher at 83.7%. Sibiński and colleagues compared 107 hips treated with preliminary traction and closed reduction with a historical cohort of 48 hips treated without traction. They reported a 42% rate of AVN in the group treated without traction compared with a 29% rate in those treated with traction. However, because of the large difference in group size between the two groups, caution should be exercised in interpreting these data. Kutlu and colleagues reported their results of 2 cohorts consisting of 89 hips treated with preliminary traction and 65 hips treated without preliminary traction and found no statistically significant differences in the rates of AVN between the two groups. Brougham and colleagues also found no significant difference in the rate of AVN between 42 hips treated without traction, 86 hips treated with traction at 60°, and 82 hips treated with traction at 90°.

There remains controversy regarding the timing of closed reduction and the relationship to the presence or absence of the ossific nucleus. Clarke and colleagues stated that before ossification the vasculature of the capital femoral epiphysis is vulnerable to compression and ischemia and, as such, reduction should be delayed until the ossific nucleus is visible because it provides mechanical strength and indicates the establishment of a collateral circulation. Segal and colleagues supported these findings with a porcine and finite element model. Segal and colleagues reported a 4% rate of AVN following reduction in the presence of an ossific nucleus compared with 53% if reduction was performed before the development of the ossific nucleus as identified on radiographs in their retrospective review of 57 hips. Clarke and colleagues proposed a treatment algorithm for delaying surgical intervention until the ossific nucleus is present radiographically. The noncontrolled series delayed treatment until the ossific nucleus was present or the patient reached 13 months of age and showed no increase in AVN compared with other series. However, Luhmann and colleagues found no difference in the rates of AVN with respect to the presence or absence of the ossific nucleus. These results were supported by the same findings from a series by Cooke and colleagues. Both series were retrospective ( Table 3 ).

Open reduction

Open reduction of the dislocated hip can be performed through various approaches. The medial approach, first described by Ludloff in 1908 and its subsequent modifications remain controversial. Advocates recommend it for the cosmetic scar, the ability to avoid damage to the abductors and the iliac apophysis, less joint stiffness, direct access to the obstacles to reduction, low blood loss, and the ability to perform bilateral surgery in 1 sitting. However, it is also criticized because there is a potential risk of AVN caused by damage to the medial circumflex vessels, the acetabulum is not easily visualized, and a capsulorrhaphy or simultaneous secondary procedures cannot be performed.

The redislocation rate for this technique varies between 0% and 6.1% ( Table 4 ), with rates of AVN varying from 8.9% to 45.4%. There is variation from 11.1% to 85.4% in the reported rates of reoperation. There is no clear association between ligation of the medial femoral circumflex vessels and the rate of AVN.

Compared with the evidence for the medial approach to reduction, less has been published on the anterior or Smith-Peterson approach, likely because it is easier to combine the anterior approach with concomitant osteotomies compared with the medial approach. Doudoulakis and Cavadias reported a series of 69 hips reduced using the Smith-Peterson approach and found 1 redislocated hip, an AVN rate of 13%, and a reoperation rate of 18 cases (26%). Dhar and colleagues reported a series of 99 hips treated by open reduction via a Smith-Peterson approach. The rate of AVN was 23.2%, with a redislocation rate of 1.8% and 24 further operative procedures. In their series of 113 hips, Szepesi and colleagues found a redislocation rate of 2.7% (3 hips), a 21.2% rate of further procedures, and a rate of AVN of approximately 11%. At a mean follow-up of 18 years, Varner and colleagues reported 82% excellent, 3% good, 10% fair, and 5% poor Iowa hip scores following anterior open reduction and, if required, femoral or acetabular osteotomy. Cordier and colleagues reported on the Tonnis anterior approach in 118 hips with a redislocation rate of 11% and AVN rate of 6%.

There is conflicting evidence regarding the timing of medial open reduction in relation to the age of the patient. Okano and colleagues found unacceptable radiological results and AVN rates in patients more than 17 months of age. In contrast, Altay and colleagues found no correlation between age at time of surgery and rate of AVN.

Pelvic osteotomies

Multiple pelvic and periacetabular osteotomies have been described for the treatment of DDH. They can be classified into reorienting osteotomies, volume-reducing osteotomies, and salvage procedures.

Salter first described the innominate osteotomy to redirect and stabilize the dislocated hip in children aged 18 months to 6 years or to treat residual dysplasia or recurrent dislocation in 1961. Pemberton reported his pericapsular osteotomy in 1965. This procedure is an incomplete transiliac osteotomy hinged about the triradiate cartilage that has the ability to improve both anterior and lateral coverage of the femoral head. Böhm and colleagues reported the long-term outcome of the Salter innominate osteotomy (SIO), noting that increasing severity of dislocation was associated with worse outcome, with a possible increase in the risk of AVN if the SIO and open reduction were performed concomitantly. Thomas and colleagues reported on the 45-year follow-up of Salter’s series of patients undergoing surgery between 18 months and 4.7 years and found that bilateral hip dislocation and postoperative complications were negative predictive factors and that there was an overall failure rate, defined as total joint arthroplasty, of 46%. Age at the time of index surgery did not affect outcome, although the investigators conceded that this may be because of type II error. Barrett and colleagues found no difference in outcome for patients treated with a concomitant open reduction with those treated with a prior open reduction and an SIO at a later stage. However, there was a worse outcome for those patients more than 4 years of age. Haidar and colleagues had similar results in their series ( Fig. 2 ).

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Trauma Pediatrics Orthopedics Shoulder Instability in Older Patients Component Selection in Revision Total Hip Arthroplasty Upper Extremity Use of Bisphosphonates in Orthopedic Surgery

Stay updated, free articles. Join our Telegram channel

Join