Journal Logo


Pelvic Osteotomies for the Treatment of Hip Dysplasia in Children and Young Adults

Gillingham, Bruce L. MC, USN; Sanchez, Anthony A. MC, USNR; Wenger, Dennis R. MD

Author Information
Journal of the American Academy of Orthopaedic Surgeons: September 1999 - Volume 7 - Issue 5 - p 325-337
  • Free


Although the true incidence of persistent acetabular dysplasia in adults is unknown, its role as a significant cause of premature hip osteoarthritis is well established.1–4 It is estimated that between 20% and 50% of cases of degenerative disease of the hip are secondary to subluxation or acetabular dysplasia.3 In a classic study of 130 patients with degenerative arthritis of the hip, Stulberg and Harris2 found that 63 patients (48%) had underlying acetabular dysplasia.

Ideally, patients with acetabular dysplasia and subluxation are identified and treated in infancy. Failing this, treatment instituted as early as possible in childhood, preferably before age 4, will take maximum advantage of the inherent remodeling capabilities of the hip joint.5 In the older child or young adult, recognition of persistent acetabular dysplasia may at least allow treatment before the onset of irreversible cartilage injury and thereby favorably influence an otherwise worrisome natural history.

The goal of treatment of patients with persistent acetabular dysplasia is to forestall or prevent the development of osteoarthritis and to obviate the need for arthroplasty at a relatively young age. Pelvic osteotomies can play a central role in this strategy by reorienting the architecture of the pelvis so as to normalize the forces of weight bearing. To enhance understanding of this important tool in the prevention of hip osteoarthritis, we will outline the general categories of pelvic osteotomies, discuss the indications for their use, and provide a brief technical description of those that are most commonly utilized in current practice.

Etiology and Biomechanics of Acetabular Dysplasia

The abnormally steep acetabular roof and the shallow joint surface that characterize acetabular dysplasia can be due to several causes. Although acetabular dysplasia is most commonly seen as a component of developmental dysplasia of the hip (DDH), it can also be a residual of Legg-Perthes disease and is frequently seen in patients with neuromuscular diseases, such as cerebral palsy and myelomeningocele.

The importance of a concentric hip reduction during development of both components of the hip joint is well known.5,6 The development of appropriate acetabular depth depends in large part on the stimulus of the femoral head pushing into the triradiate cartilage. Conversely, if the femoral head is to achieve its normal spherical shape in adulthood, it must be well seated within the acetabulum during infancy and early childhood. Any condition that interferes with this interdependent relationship can lead to acetabular dysplasia. Left untreated, a meaningful reduction in the longevity of the hip joint can result.2

When evaluating an infant or child with acetabular dysplasia, it is important to ascertain whether there is associated hip subluxation. Disruption of Shenton's line on a standing anteroposterior (AP) radiograph of the pelvis indicates that the femoral head is proximally and laterally subluxated. (Shenton's line is created by drawing a line along the proximal medial femoral metaphysis and extending onto the superior border of the obturator foramen. In a normal hip, the line is an arc of continuous contour.) If the condition is not treated, premature joint degeneration and clinical disability predictably follow. The age at onset of symptoms correlates well with the severity of the subluxation.2,4,7 The patients with the most serious disease begin to experience pain in the second decade; those with the least serious disease may reach the fifth decade before noting symptoms.7 However, even in moderate cases, functionally disabling symptoms can occur during the most vigorous and productive years of life.

Less predictable is the effect of acetabular dysplasia without subluxation. Patients without subluxation are frequently asymptomatic, and osteoarthritis tends to develop much later and less commonly than in patients with subluxation.1 Cooperman et al1 concluded that the usual radiographic indices by which acetabular morphology is characterized are not predictive of the rate at which osteoarthritis develops in patients with acetabular dysplasia alone. In a more recent study of the contralateral hip in patients who had undergone a total hip replacement for osteoarthritis secondary to dysplasia, Murphy et al8 noted a significant (P<0.0001) difference in standard radiographic indices between patients in whom osteoarthritis developed in the opposite hip before age 65 and those in whom it did not. Adolescent patients with acetabular dysplasia without associated subluxation should be carefully evaluated for clinical findings, as they (particularly females) are at greater risk than the general population for the development of clinical signs and symptoms of degenerative hip disease.2

Acetabular dysplasia eventually results in degenerative joint disease due to alterations in load. In a study of patients followed up for an average of 29 years after reduction of a congenitally dislocated hip, Hadley et al9 concluded that over the long term, articular cartilage can tolerate mean contact pressures of 2 MPa, with an inverse relationship between higher pressures and the time needed to develop degenerative changes. In a normal hip, peak acetabular contact pressures may be as high as 10 MPa, depending on activity, but are distributed throughout the entire acetabulum, with the dome being subjected to the highest pressure and the rim to essentially none.10

In the dysplastic hip, point loading occurs at the edge of the steep, shallow acetabulum. Pelvic osteotomies reduce this load by increasing contact area, relaxing the capsule and muscles about the hip, and improving the moment arm of the hip.10

Clinical Evaluation

History and Physical Examination

The initial evaluation focuses on a characterization of the patient's symptoms, activity level, functional limitations, and expectations. The patient with residual dysplasia may be asymptomatic or may experience only vague discomfort with strenuous weight-bearing activities. The presence of more extensive complaints may suggest that degenerative arthritis is already established. Most children and teenagers with radiographic evidence of severe subluxation that requires treatment will have no symptoms.

Physical examination includes inspection of the extremity for muscle wasting of the quadriceps and gluteus, longitudinal malalignment of the limb, and limb-length discrepancy. Analysis of the patient's gait may reveal a limp or an abductor lurch. Palpation of the posterior iliac crest is performed with the patient standing, in order to detect pelvic obliquity. The Trendelenburg test, performed both immediately and after a 20-second delay, allows assessment of the competence and resistance to fatigue of the hip abductors. The range of hip motion is carefully documented, with particular attention paid to hip rotation and the presence of contractures. Pain with active or passive motion suggests synovitis or impingement.

Pain elicited by internal rotation of the flexed and adducted hip may signify the presence of a detached limbus. Termed “acetabular rim syndrome” by Klaue et al,11 this condition is considered to be a precursor of osteoarthritis of the hip in patients with acetabular dysplasia. Most children and adolescents, even those with severe dysplasia, have few, if any, physical findings.

Radiologic Evaluation

A comprehensive radiographic evaluation is essential to clarify the degree of deformity before performing a hip osteotomy. The initial plain films should include weight-bearing (standing) AP, frogleg lateral, and abduction-internal rotation views of the pelvis and hips, as well as a faux profil (French for “false profile”) view of the affected hip.

Analysis of the AP view should include a qualitative assessment of Shenton's line and the acetabular sourcil (a term derived from the French word for “eyebrow”). As mentioned previously, a break in the normal smooth arc of Shenton's line is suggestive of hip subluxation.

The appearance of the acetabular sourcil is a sensitive radiographic indicator of asymmetric loading of the hip joint.12 Ordinarily, this dense subchondral bone appears as a smooth curve of uniform thickness. In the dysplastic hip, lateral sourcil thickening occurs, which represents increased focal loading due to underlying malalignment (Fig. 1). Careful analysis of the sourcil, especially if there is side-to-side asymmetry, is an important tool for the hip surgeon.

Fig. 1
Fig. 1:
Anteroposterior radiograph of the pelvis of an 18-year-old girl with left hip dysplasia. The right acetabular sourcil appears as a smooth curve of uniform thickness. In contrast, the left sourcil is wider laterally, indicating focal loading.

Measurements of the centeredge angle of Wiberg,4 the adult acetabular angle of Sharp,13 and the acetabular depth are useful means of quantitating the severity of the dysplasia. Prior to ossification of the femoral head, the acetabular index is generally used to assess acetabular configuration, with normal mean values decreasing to less than 20 degrees by age 24 months (Table 1).14

Table 1
Table 1:
Radiographic Assessment of the Hip*

The abduction-internal rotation view neutralizes femoral anteversion and allows accurate assessment of the true femoral neck-shaft angle. In addition, this view simulates the coverage possible with a proximal femoral varus osteotomy. A single innominate osteotomy can be simulated by taking a radio- graph with the hip held in 25 degrees of flexion, 10 degrees of abduction, and neutral rotation, with the x-ray beam directed posteriorly and caudally 25 degrees.15

The faux profil view, first described by Lequesne and de Sèze16 in 1961, is a true lateral view of the acetabulum made with the patient standing with the pelvis rotated 25 degrees toward the x-ray beam. The result is a centeredge angle-type assessment of anterior coverage.

Fluoroscopic examination of the hip in various positions can be a useful means of gauging the effect of a proposed osteotomy. Arthrography is of particular value when assessing the incompletely ossified femoral head of a child.

The recent advances in three-dimensional reconstruction of computed tomographic (CT) studies have greatly expanded our understanding of the underlying pathoanatomy in patients with residual acetabular dysplasia.

This modality is emerging as a valuable preoperative tool for surgical decision making and planning.17,18 The three-dimensional display provides superior information about the fit of the femoral head in the acetabulum, as well as the size, shape, and orientation of the acetabulum. This increased detail has clearly demonstrated that acetabular dysplasia is more complex than previously thought. More than just malrotation or maldirection, acetabular dysplasia is a combination of acetabular maldirection, margin erosion, torsion, hypoplasia (localized or global), abnormal shape, and decreased acetabular surface area.19 With this greater understanding has come an improved ability to match the osteotomy to the type of acetabular deficiency. In addition, comparison of the initial study with one performed postoperatively is a valuable tool in assessing the success of the osteotomy in reproducing normal hip architecture and will further refine the surgeon's ability to choose the appropriate osteotomy.

Treatment Overview

The goal of treatment of the patient with acetabular dysplasia is to establish normal biomechanical forces about the hip joint. The means by which this can be accomplished vary greatly depending on the patient's age, the severity of dysplasia, and the morphology of the hip joint (Fig. 2).

Fig. 2
Fig. 2:
Algorithm for treatment of developmental dysplasia of the hip.

In the infant and young child, the initial assessment and treatment are directed toward ensuring that a concentric hip reduction is present. Restoration of hip-joint concentricity may set the stage for profound remodeling. The amount of remodeling that can be relied on to produce a normal acetabulum and the relationship of remodeling and age have been debated. Salter stated that the potential for remod- eling was greatest at birth and gradually decreased until age 18 months, when it would no longer be

ensured.20 Harris5 concluded, on

the basis of a prospective radiologic study of 79 dislocated hips, that the critical point beyond which restoration of hip congruence would not necessarily result in a normal acetabulum was age 4; the risk of producing a moderately or severely dysplastic acetabulum more than doubled if hip reduction occurred beyond this age. Thus, in the 18-month-old to 3-year-old child with acetabular dysplasia due to hip dislocation, the need for an acetabular procedure at the time of open reduction and capsulorrhaphy is controversial. One approach is to delay a secondary procedure so that the adequacy of the acetabular remodeling response can be assessed.21 Most North American centers, however, are moving toward following Salter's advice to include the acetabular procedure at the time of primary treatment in order to maximize the likelihood that a normal acetabulum will ultimately develop.20,22

DelBello et al22 compared the results of open reduction alone, open reduction followed by delayed innominate osteotomy, and open reduction plus immediate innominate osteotomy in a group of older children with DDH. Only the immediate osteotomy group achieved acetabular indices comparable to those in normal control subjects. In addition, the hips of 95% of the patients in the immediate innominate osteotomy group were classified as group I or II on the Severin hip dysplasia scale, compared with 61% in the group of patients who underwent open reduction alone and 60% in the delayed innominate osteotomy group. No differences were found in the rate of osteonecrosis of the femoral head, the estimated surgical blood loss, and the operative time. On the basis of these findings, the authors recommended routine innominate osteotomy at the time of open reduction for all patients over the age of 18 months.

In the child aged 3 years or older, innominate osteotomy is performed routinely because of the unpredictable remodeling potential beyond this age.23 In the patient older than age 8, the treatment plan is based on the symptoms and the severity of the residual dysplasia. In terms of clinical characteristics, patients with DDH vary greatly, ranging from the asymptomatic patient in whom the dysplasia is an incidental finding to the severely incapacitated individual who is in constant pain.

Nonoperative therapy is directed toward minimizing excessive point loading across the joint by avoiding unnecessary impact-loading activities, achieving ideal body weight, and maintaining overall physical fitness and muscle tone. This can be supplemented by judicious use of nonsteroidal anti-inflammatory agents in patients for whom they are not contraindicated. While this can be an effective means of treating the symptoms, this strategy does not address the underlying malalignment that is the ultimate cause of osteoarthritis. It is important that parents and older patients understand this, particularly when there are few, if any, symptoms. Paradoxically, these are the patients who may have the most to gain from surgery.

The goal and type of surgery chosen depend on the severity of the patient's condition. In general, there are two groups of patients with residual dysplasia.24 The group amenable to “reconstructive” procedures are minimally symptomatic and have normal joint congruence and no irreversible cartilage injury. The goal of surgery in this setting is realignment of the joint surfaces to produce more nor mal loading and to forestall the development of osteoarthritis. Reconstructive osteotomies include Salter, Pemberton, triple innominate, and periacetabular procedures.20,25–27 These procedures have a relatively predictable outcome.

In contrast, the “salvage” group demonstrate evidence of irreversible cartilage injury. In this setting, the goal of surgery is to relieve pain, delay the inevitable arthroplasty, and improve function in the meantime. The procedures used have a less certain outcome. Examples include Chiari28 and shelf procedures, often performed on an incongruent, already distorted hip joint.

Primary Treatment of Complete Hip Dislocation

If closed treatment has not been effective in a child less than 18 months of age, establishment of normal hip alignment can generally be achieved by open reduction and capsulorrhaphy alone. In patients between 18 months and 3 years of age, immediate innominate osteotomy at the time of open reduction maximizes the remodeling potential of the hip. In the presence of excessive femoral anteversion or valgus of the femoral neck, a varus derotational osteotomy can be included (Fig. 3).

Fig. 3
Fig. 3:
Bilateral DDH in a 2-year-old girl. A, Preoperative AP radiograph of the pelvis. B, AP radiograph obtained after staged bilateral open reductions, proximal femoral varus osteotomies, and innominate osteotomies. Ideal graft position is present on the left; undesirable slightly posterior displacement of the osteotomy is present on the right.

We do not advocate the Pemberton or Dega procedure as part of the treatment for primary dislocation of the hip because both either decrease acetabular volume or change acetabular shape. The Salter procedure allows redirection but avoids shape or volume changes.

For the child over age 3 years with untreated DDH, the approach is similar. The only difference is the addition of a derotational femoral shortening osteotomy to decrease the compressive forces on the femoral head after reduction.

Treatment of Residual Dysplasia

Treatment of residual dysplasia is based on the patient's age and the presence or absence of congruent hip reduction. In the child between the ages of 2 and 10 with a well-reduced hip with anterolateral acetabular deficiency of a moderate degree, either a Salter or a Pemberton procedure can be performed. Theoretically, with the Pemberton procedure, the relative relationship of the cuts in the inner and outer tables of the ilium can be adjusted to change the location of the desired augmentation. As a configuration-changing osteotomy, the Pemberton is particularly helpful in the case of a shallow, capacious, or “wandering” acetabulum.

Although many cogent arguments have been made regarding the relative usefulness of the Salter and Pemberton procedures, there is little objective basis for choosing one over the other. The Pemberton procedure has the advantage that because the osteotomy is so secure, fixation pins are not required, and a second operation to remove pins is unnecessary. If severe coxa valga or anteversion is present, a femoral osteotomy may have to be added to achieve an anatomically correct radiographic appearance.

In the older child or adolescent in whom the triradiate cartilage remains open (usually over age 10), the triple innominate osteotomy26 is our procedure of choice. Although a more extensive dissection is required, this procedure offers the advantage of increased acetabular fragment mobility and thus a wider range of coverage options.

After triradiate cartilage closure, the periacetabular osteotomy popularized by Ganz et al27 can be considered in addition to a triple osteotomy. The Ganz procedure is technically challenging, but it affords nearly unlimited acetabular mobility.

In the patient between 2 and 8 years of age with an incongruent joint, a Salter or Pemberton innominate osteotomy can be considered provided sufficient remodeling potential is present. However, the outcome is less predictable than in the child less than age 4 with concentric reduction. In such circumstances, a femoral osteotomy is commonly added to decompress the joint. Beyond age 8, when no remodeling will occur, a salvage osteotomy, such as a Chiari or shelf procedure, is necessary.28,29

Redirectional (Complete) Osteotomies

As their name implies, redirectional osteotomies improve the coverage of the femoral head by shifting the position of the acetabulum. The acetabulum itself remains unchanged. Because these procedures involve complete cuts through the innominate bone, fixation is required to maintain the new alignment until the osteotomy heals. In general, there is a direct relationship between the technical complexity of a redirectional osteotomy and the amount and range of coverage possible.

Single Innominate Osteotomy (Salter)

First described by Salter in 1961,20 the single innominate osteotomy remains in widespread use (Fig. 4). Salter conceptualized that acetabular maldirection was responsible for the deficiency of femoral head coverage and designed an osteotomy to redirect the acetabulum and thereby provide a stable hip reduction in the functional weight-bearing position. The amount of acetabular fragment mobility obtained with the single innominate osteotomy depends on rotation and hinging through the pliable fulcrum of the pubic symphysis.

Fig. 4
Fig. 4:
Salter single innominate osteotomy.

The primary indication for a Salter osteotomy is a deficiency of anterolateral femoral head coverage in an otherwise concentrically reduced hip. A shallow acetabulum is a relative contraindication. Improvement in the center-edge angle of 20 to 22 degrees and a 10-degree improvement in the acetabular index can be expected.30,31

Skillful exposure and clearing of the sciatic notch is critical. The Gigli saw cut begins low in the sciatic notch and exits just above the anterior inferior iliac spine. The inferior fragment is pulled distally and anteriorly with a towel clamp to provide increased anterolateral coverage. Care should be taken to prevent displacing the proximal fragment, as this will lead to opening of the osteotomy posteriorly at the sciatic notch, resulting primarily in leg lengthening. Mobility of the fragment is enhanced by placing the ipsilateral leg into the figure-of-four position of hip flexion, abduction, and external rotation. A triangular wedge of anterior iliac crest is fixed into place at the osteotomy site with two threaded pins directed posteromedially. These are subsequently removed at 6 to 8 weeks in young children and at 3 to 4 months in older children, who are at a greater risk for graft collapse. An iliopsoas tenotomy at the pelvic brim is considered an essential component of the procedure to both decrease compression on the femoral head and allow distal fragment mobility.

In a review of the 15-year data on 140 patients treated primarily for hip dislocation and subluxation, Salter and Dubos32 reported 93.6% excellent or good results in patients aged 18 months to 4 years, with no failures. In the 4- to 10-year-old age group, however, only 56.7% had good or excellent results, and the failure rate was 6.6%. Initial operative complications included superficial wound infections (incidence of 1.5%), femoral head osteonecrosis (5.7%), loss of osteotomy position prior to the use of two pins (2.8%), redislocation (5.6%), resubluxation (14.3%), and supracondylar femur fracture (6.4%). There were no deep infections. Therefore, because less satisfactory results generally occur in older children, triple innominate osteotomy is preferable for children over age 10.

Triple Innominate Osteotomy

The triple osteotomy described by Steel26 in 1965 consists of osteotomies of the ischium and pubis in addition to a Salter innominate osteotomy (Fig. 5). Tšnnis et al33 have described a modification of the triple osteotomy that is also widely used. The triple osteotomy is generally indicated for older children and adolescents, in whom symphyseal rotation is more limited than in younger children due to skeletal maturity and decreased ligamentous laxity (Fig. 6). As is the case with the single innominate osteotomy, concentric hip reduction is a prerequisite. Because it is a circumacetabular osteotomy, significant mobility of the acetabular fragment is possible. In addition, if the ischial osteotomy is directed obliquely from lateral to medial, the acetabular fragment can be displaced medially, moving the hip center to a more physiologic position and improving gait mechanics.

Fig. 5
Fig. 5:
Triple innominate osteotomy.
Fig. 6
Fig. 6:
Images of a female patient with left acetabular dysplasia (same patient as in Fig. 1). A, At age 18, AP radiograph of the abducted, internally rotated left hip shows joint congruence. B, AP view obtained at age 19, 11Ú2 years after left proximal femoral varus osteotomy. C, Three-dimensional reconstruction of a CT study of the hip obtained at age 21 demonstrates persistent left acetabular dysplasia with deficient anterolateral coverage. D, AP radiograph obtained after triple innominate osteotomy on the left shows improved coverage.

The triple osteotomy requires two incisions. Steel26 originally advised a horizontal incision made over the ischial tuberosity for the ischial osteotomy and an anterior incision (as in the single innominate procedure) for the pubic and innominate osteotomies. We now prefer to make the Salter cut only through the anterolateral incision and to make the superior pubic ramus and ischial cuts through a single transverse groin incision. Placement of bone graft and fixation are performed as for the Salter osteotomy. Care must be taken to avoid excessive external rotation of the free distal fragment. Also, muscle should not intervene in the pubic and ischial cuts, as pseudarthrosis can occur. Use of a temporary Schanz screw to guide the acetabular segment allows appropriate acetabular redirection, avoiding excessive external rotation.

In Steel's original series,26 the results were considered to be satisfactory in 19 of the 23 hips with congenital dysplasia followed up between 2 and 10 years postoperatively. Five complications were reported: two cases of postoperative ileus, one ischial wound infection (Escherichia coli), and two instances of pressure necrosis of the skin over the anterior inferior spine of the displaced acetabular fragment.

In a review of 44 patients (56 hips) an average of 7 years after triple osteotomy, Faciszewski et al34 reported improvement in both pain and function in 94% of patients. The center-edge angle increased by a mean of 33 degrees, and the acetabular angle decreased by a mean of 15 degrees. Two complications occurred: one pulmonary embolism and one superficial wound infection.

Periacetabular Osteotomy

Introduced in 1988 by Ganz et al,27 the periacetabular osteotomy allows extensive acetabular reorientation, including medial and lateral displacement. Osteotomies are performed in the pubis, ilium, and ischium. A vertical posteriorcolumn osteotomy connects the posterior extremes of the iliac and ischial osteotomies approximately 1 cm anterior to the sciatic notch (Fig. 7). This osteotomy must be done after skeletal maturity, because it crosses the triradiate cartilage; therefore, it is indicated only for older adolescent and adult patients with dysplastic hips that require improvement of congruency and containment (Fig. 8).

Fig. 7
Fig. 7:
Ganz periacetabular osteotomy.
Fig. 8
Fig. 8:
Images of a 15-year-old girl with left acetabular dysplasia. A, Preoperative AP radiograph of the pelvis. B, Three-dimensional CT reconstruction shows significant uncovering of the left hip. C, AP radiograph obtained after a Ganz periacetabular osteotomy.

Because no complete cut is made into the sciatic notch (the posterior column is split vertically), the Ganz procedure is very stable, and no postoperative cast is required. Immediate crutch weight bearing is advised. Additional advantages include preservation of the blood supply to the acetabular fragment, use of a single surgical approach, and preservation of the shape of the pelvis, which permits normal vaginal delivery. A disadvantage is that the procedure is difficult to learn.

In 1995, Trousdale et al35 reported the results of use of the Ganz periacetabular osteotomy, with or without intertrochanteric osteotomy, on 42 patients with dysplastic, osteoarthritic hips. The average follow-up interval was 4 years. They found that the Harris hip score improved from an average of 62 points preoperatively to 86 points postoperatively (P<0.0001). The average anterior-edge angle was 1 degree preoperatively and 27 degrees postoperatively. Complications included deep venous thrombosis in 2 patients, pubic nonunion in 2, and heterotopic bone formation in 14. There were no major neurovascular complications in this series, although nerve and/or vessel injury is possible with this rather complex osteotomy.

Reshaping (Incomplete) Osteotomies

The object of reshaping osteotomies is restoration of acetabular morphology by changing the shape of the acetabulum. They are primarily indicated for the patient with a capacious or wandering acetabulum, as is commonly seen in both childhood DDH and cerebral palsy. Depending on the osteotomy, additional lateral and/or posterior coverage can be obtained. They should be used only in the skeletally immature, as the osteotomy hinges through the triradiate cartilage. Because the osteotomy is incomplete, there is inherent stability after graft placement, and internal fixation is not required.

Pemberton Osteotomy

In 1965, Pemberton25 described an incomplete osteotomy that hinges through the triradiate cartilage (Fig. 9). A Smith-Peterson approach is used to expose the inner and outer tables of the ilium. The outer ilium is osteotomized beginning immediately superior to the anterior inferior iliac spine. The osteotomy extends posteriorly 0.25 inch above and parallel to the joint capsule and is carried to the ilioischial limb of the triradiate cartilage. The inner ilium is osteoto-mized separately to match the outer cut (if anterior coverage is primarily desired); alternatively, the posterior limb may be shifted anteriorly to provide additional lateral coverage. Care must be taken to remain halfway between the anterior edge of the sciatic notch and the posterior rim of the acetabulum.

Fig. 9
Fig. 9:
Pemberton osteotomy.

The osteotomy is opened enough to create an acetabular angle of approximately 0 degrees. A groove is created to hold the graft in position, and a triangular wedge of bone from the anterior ilium is placed and impacted. No internal fixation is required. A spica cast is used for 8 weeks. This osteotomy is appropriate for patients between the age of 18 months and skeletal maturity (Fig. 10).

Fig. 10
Fig. 10:
Images of a girl with bilateral acetabular dysplasia. A, At 3 years of age, AP radiograph of the pelvis showed bilateral dysplasia. B, At 71Ú2 years of age, three-dimensional reconstruction of a CT study of the pelvis showed persistent dysplasia. C, Radiograph of the right hip obtained during a Pemberton osteotomy. D, Anteroposterior radiograph of the pelvis obtained 2 months after bilateral Pemberton osteotomies shows graft incorporation and improved coverage.

In Pemberton's original series of 91 patients (115 hips), all 46 children less than 4 years of age had a good result.25 There were 20 good and 4 fair results in children aged 4 to 7 years, and 12 good, 6 fair, and 3 poor results in children between the ages of 7 and 12 years. Com- plications included 5 redislocations, 12 cases of osteonecrosis (all eventually healed), and 2 sciatic nerve palsies (in cases of high dislocation).

Faciszewski et al36 reported the results in 52 hips with acetabular dysplasia treated with the Pemberton osteotomy. The average age of the 42 patients was 4 years at the time of surgery. The mean follow-up interval was 10 years. At the time of follow-up, 42 hips (81%) were radiographically normal. No patient had pain, three patients had decreased range of motion, and one patient had a positive Trendelenburg test. There were no complications.

Dega Osteotomy

The Dega osteotomy is an acetabuloplasty that also changes ace- tabular configuration and shape. It provides increased posterolateral coverage by means of an osteotomy of the lateral cortex of the ilium only, hinging through the open triradiate cartilage.37 The primary indication for this osteotomy is the presence of a capacious acetabulum with posterolateral deficiency, as is often found in children with cerebral palsy. Some authors also use it for persistent acetabular dysplasia in DDH. The technique has been well described by Mubarak et al.37

Salvage/Augmentation Procedures

A salvage osteotomy may be indicated when a congruent reduction between the femoral head and the acetabulum cannot be obtained. Such an osteotomy may also be appropriate for an adolescent or adult with a painful subluxated hip or for a patient with prior surgical failures. Selection of one of the various salvage/augmentation procedures presupposes that the remaining hyaline cartilage is inadequate or cannot be redirected to provide coverage of the femoral head. These procedures act to increase the weight-bearing surface and rely on capsular metaplasia to provide an articulating surface. Commonly, an associated valgus or varus osteotomy will help to realign the joint and decompress the joint.

Chiari Osteotomy

The primary indication for use of the Chiari osteotomy is a painful, subluxated hip without the possi- bility of congruent reduction in a patient older than 8 years of age. Prerequisites include a satisfactory range of motion, maintenance of the cartilage space, and minimal osteoarthritis.38 The amount of coverage obtained depends on the width of the ilium. In addition to widening the contact area of the femoral head, the hip joint is medialized, reducing the workload of the hip.28

The Chiari osteotomy is performed through an anterolateral approach. The inner and outer tables of the ilium are exposed to the sciatic notch. The ilium is osteotomized between the insertion of the capsule and the reflected tendon of the rectus. The osteotomy follows the capsular insertion in a curved line to the sciatic notch posteriorly and the inferior iliac spine anteriorly (Fig. 11). The inner table of the ilium is cut at a point slightly more superior than the outer table. The acetabulum is displaced medially by forced abduction with rotation through the pubic symphysis, such that the femoral head is covered without completely separating the fragments. Supplemental shelf augmentation is often required to provide adequate coverage. The osteotomy is fixed internally, and a hip spica cast is applied for 4 to 6 weeks.28

Fig. 11
Fig. 11:
Chiari osteotomy.

Chiari28 reported good or excellent results in 66% of his initial 200 cases in patients under 16 years of age, who were followed up for 2 to 8 years. The complications included nerve palsy in eight patients and restricted range of motion in two.

Windhager et al38 reported the results of 236 Chiari osteotomies (more than 90% of which were done by Chiari) followed for an average of 24.8 years. Overall, the results of 52% were clinically rated excellent or good; 30%, fair; and 18%, poor. Results worsened with increasing age and in the presence of preoperative degenerative joint disease. Complications included incomplete medialization (4 hips), transient peroneal nerve palsy (2 hips), transient meralgia paresthetica (1 hip), infection (5 hips), limited range of motion requiring manipulation (4 hips), and heterotopic ossification (1 hip).

Slotted-Shelf Procedure

Shelf procedures create an extraarticular buttress, preventing further subluxation while increasing the load-bearing area of the hip. They are indicated when a congruent reduction is impossible, when a patient has no significant degenerative joint disease, and when augmentation is needed after other osteotomies.

The slotted-shelf procedure is preferred to other types of shelf procedures because of the greater resultant stability. As described by Staheli and Chew,29 the procedure involves an iliofemoral approach to expose the hip joint, followed by division of the reflected head of the rectus femoris. The hip capsule is thinned to approximately 6 to 7 mm by “filleting.” The acetabular slot is placed exactly at the acetabular margin, with the floor of the slot consisting of the articular cartilage of the acetabulum. A 1-cm-deep slot is made and continued as far anteriorly and posteriorly as is needed for coverage without blocking flexion anteriorly.

Thin strips of corticocancellous bone are taken from the lateral ilium. The strips are placed radially into the slot, followed by a second layer placed at right angles to the first (Fig. 12). Both layers are then secured by bringing the reflected head of the rectus forward over the graft and suturing it back to its original position. The remaining graft is placed above the initial layer to act as an additional buttress. A hip spica cast is worn for 6 weeks.29

Fig. 12
Fig. 12:
Slotted-shelf augmentation.

Staheli and Chew29 reported the results of slotted-shelf acetabular augmentation procedures in 108 hips in 98 children at an average of 5 years (24 to 168 months). The results were good or excellent in 90 hips (83.3%). Complications occurred in 26 hips and included placement of a graft that was too wide (9 hips) or too narrow (7 hips), delayed union or nonunion (3 hips), dysesthesia of the lateral femoral cutaneous nerve (4 hips), persistent pain (2 hips), and femur fracture (1 hip).


Persistent acetabular dysplasia is a well-known cause of premature hip osteoarthritis. Every attempt should be made to identify and treat infants and young children whose hips are not concentrically reduced in order to ensure the development of a normal hip joint that will last a lifetime. Ideally, corrective osteotomies should be done by age 4 to 5 years to achieve an almost completely normal hip. In adolescent and young adult patients with persistent acetabular dysplasia, aggressive treatment to restore normal hip architecture is indicated to prevent irreversible cartilage injury. Proper selection and performance of a pelvic osteotomy provides a biologic solution, utilizing the patient's own tissues to favorably influence what might otherwise be a worrisome natural history.


1. Cooperman DR, Wallensten R, Stulberg SD: Acetabular dysplasia in the adult. Clin Orthop 1983;175:79-85.
2. Stulberg SD, Harris WH: Acetabular dysplasia and development of osteoarthritis of hip, in Harris WH (ed): The Hip: Proceedings of the Second Open Scientific Meeting of the Hip Society, 1974. St Louis: Mosby, 1974, pp 82-93.
3. Weinstein SL: Natural history of congenital hip dislocation (CDH) and hip dysplasia. Clin Orthop 1987;225:62-76.
4. Wiberg G: Studies on dysplastic acetabula and congenital subluxation of the hip joint: With special reference to the complication of osteoarthritis. Acta Chir Scand Suppl 1939;58:7-135.
5. Harris NH: Acetabular growth potential in congenital dislocation of the hip and some factors upon which it may depend. Clin Orthop 1976;119:99-106.
6. Ponseti IV: Growth and development of the acetabulum in the normal child: Anatomical, histological, and roentgenographic studies. J Bone Joint Surg Am 1978;60:575-585.
7. Wedge JH, Wasylenko MJ: The natural history of congenital dislocation of the hip: A critical review. Clin Orthop 1978;137:154-162.
8. Murphy SB, Ganz R, MŸller ME: The prognosis in untreated dysplasia of the hip: A study of radiographic factors that predict the outcome. J Bone Joint Surg Am 1995;77:985-989.
9. Hadley NA, Brown TD, Weinstein SL: The effects of contact pressure elevations and aseptic necrosis on the longterm outcome of congenital hip dislocation. J Orthop Res 1990;8:504-513.
10. Brand RA: Hip osteotomies: A biomechanical consideration. J Am Acad Orthop Surg 1997;5:282-291.
11. Klaue K, Durnin CW, Ganz R: The acetabular rim syndrome: A clinical presentation of dysplasia of the hip. J Bone Joint Surg Br 1991;73:423-429.
12. Pauwels F; Furlong RJ, Maquet P (trans): Biomechanics of the Normal and Diseased Hip: Theoretical Foundation, Technique and Results of Treatment—An Atlas. New York: Springer, 1976, p 4.
13. Sharp IK: Acetabular dysplasia: The acetabular angle. J Bone Joint Surg Br 1961;43:268-272.
14. Scoles PV, Boyd A, Jones PK: Roentgenographic parameters of the normal infant hip. J Pediatr Orthop 1987; 7:656-663.
15. Rab GT: Preoperative roentgenographic evaluation for osteotomies about the hip in children. J Bone Joint Surg Am 1981;63:306-309.
16. Lequesne M, de Sèze S: Le faux profil du bassin. Rev Rhum 1961;28:643-652.
17. Kim HT, Wenger DR: Location of acetabular deficiency and associated hip dislocation in neuromuscular hip dysplasia: Three-dimensional computed tomographic analysis. J Pediatr Orthop 1997;17:143-151.
18. Klaue K, Wallin A, Ganz R: CT evaluation of coverage and congruency of the hip prior to osteotomy. Clin Orthop 1988;232:15-25.
19. Wedge JH: Hip joint acetabular dysplasia [editorial]. J Pediatr Orthop 1997;17:141-142.
20. Salter RB: Innominate osteotomy in the treatment of congenital hip dislocation and subluxation of the hip. J Bone Joint Surg Br 1961;43:518-539.
21. Marafioti RL, Westin GW: Factors influencing the results of acetabuloplasty in children. J Bone Joint Surg Am 1980;62:765-769.
22. DelBello DA, Nattrass GR, Moseley CF, Watts HG: Acetabular development after open reduction in developmental dysplasia of the hip: The role of concurrent innominate osteotomy. Orthop Trans 1995;19:298.
23. Barrett WP, Staheli LT, Chew DE: The effectiveness of the Salter innominate osteotomy in the treatment of congenital dislocation of the hip. J Bone Joint Surg Am 1986;68:79-87.
24. Millis MB, Poss R, Murphy SB: Osteotomies of the hip in the prevention and treatment of osteoarthritis. Instr Course Lect 1992;41:145-154.
25. Pemberton PA: Pericapsular osteotomy of the ilium for treatment of congenital subluxation and dislocation of the hip. J Bone Joint Surg Am 1965;47:65-86.
26. Steel HH: Triple osteotomy of the innominate bone. J Bone Joint Surg Am 1973;55:343-350.
27. Ganz R, Klaue K, Vinh TS, Mast JW: A new periacetabular osteotomy for the treatment of hip dysplasias: Technique and preliminary results. Clin Orthop 1988;232:26-36.
28. Chiari K: Medial displacement osteotomy of the pelvis. Clin Orthop 1974; 98:55-71.
29. Staheli LT, Chew DE: Slotted acetabular augmentation in childhood and adolescence. J Pediatr Orthop 1992;12: 569-580.
30. Utterback TD, MacEwen GD: Comparison of pelvic osteotomies for the surgical correction of the congenital hip. Clin Orthop 1974;98:104-110.
31. Wong-Chung J, Ryan M, O'Brien TM: Movement of the femoral head after Salter osteotomy for acetabular dysplasia. J Bone Joint Surg Br 1990;72:563-567.
32. Salter RB, Dubos JP: The first fifteen years' personal experience with innominate osteotomy in the treatment of congenital dislocation and subluxation of the hip. Clin Orthop 1974;98:72-103.
33. Tšnnis D, Behrens K, Tscharani F: A modified technique of the triple pelvic osteotomy: Early results. J Pediatr Orthop 1981;1:241-249.
34. Faciszewski T, Coleman SS, Biddulph G: Triple innominate osteotomy for acetabular dysplasia. J Pediatr Orthop 1993;13:426-430.
35. Trousdale RT, Ekkernkamp A, Ganz R, Wallrichs SL: Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthrosis in dysplastic hips. J Bone Joint Surg Am 1995;77:73-85.
36. Faciszewski T, Kiefer GN, Coleman SS: Pemberton osteotomy for residual acetabular dysplasia in children who have congenital dislocation of the hip. J Bone Joint Surg Am 1993;75:643-649.
37. Mubarak SJ, Valencia FG, Wenger DR: One-stage correction of the spastic dislocated hip: Use of pericapsular ace-tabuloplasty to improve coverage. J Bone Joint Surg Am 1992;74:1347-1357.
38. Windhager R, Pongracz N, Schšnecker W, Kotz R: Chiari osteotomy for congenital dislocation and subluxation of the hip: Results after 20 to 34 years follow-up. J Bone Joint Surg Br 1991;73: 890-895.
© 1999 by American Academy of Orthopaedic Surgeons