Dysplasia of the hip is a common cause of secondary arthritis of the hip. These cases are complex and challenging to the arthroplasty surgeon. The normal hip anatomy is distorted and surgical complications are likely. Indications for surgery are similar to standard arthroplasty with few exceptions. Preoperative planning is essential. Classification schemes are reviewed and aid in comprehending the scope of the disease. An outcomes analysis is provided for proper perspective. Alternative bearings may transform future outcomes of arthroplasty in dysplasia. The authors’ preferred surgical technique uses cementless implants and a femoral shortening osteotomy and is described in detail.
These patients are young and revision is inherently more likely.
Nonarthroplasty surgical options may need to be considered, especially with a preserved joint space.
The complication rate is high in these cases.
Preoperative planning is vital to anticipate the availability of appropriate prostheses and equipment.
Understanding the distorted anatomy is essential in evaluating and treating the dysplastic hip.
Proper intraoperative visualization is critical to understanding the anatomic deformity.
Placement of the reconstructed acetabulum is crucial and dictates the outcome of the patient.
Bone stock often is deficient but several techniques for coverage are available. Screw fixation is prudent.
Femoral shortening often is required in cases of high dislocation depending on the chosen location of the acetabulum.
Distorted anatomy increases the difficulty of these cases and must be understood and recognized.
Nerve dysfunction is a significant surgical risk, and overlengthening must be avoided.
Achieving intraoperative rotational stability of a subtrochanteric osteotomy is essential for union.
Dislocation rate is increased. Careful intraoperative trialing, increasing offset, and large head sizes with alternative bearings may decrease this risk.
Dysplasia of the hip is a common cause of secondary arthritis of the hip. The severity of the disease varies, with a spectrum from insufficient coverage of the femoral head to complete dislocation of the hip. Patients in need of surgical intervention often are young. Planning and executing the arthroplasty in such a way that the hip will last is absolutely imperative, with the understanding that the prospect of revision is likely.
As with standard hip arthroplasty, the goals of the surgery include restoration of hip biomechanics and stable implant fixation. However, these procedures can be anything but typical, with an increased risk of complications and historically poor outcomes.
The necessity of preoperative planning and obtaining quality radiographs becomes even more important in these cases. Factors that increase the complexity include distorted anatomy, deficient bone stock, and previous surgery and hardware ( Fig. 15-1 ). These patients often require specialized implants not commonly used, which may need to be requested in advance.
A clear understanding of the distorted anatomy is an essential factor in evaluating and treating the dysplastic hip. Careful preoperative planning, a meticulous history and physical examination, and thorough patient counseling preoperatively will help minimize complications and optimize the long-term outcomes in these difficult cases.
Normal development of the hip joint requires a concentric relationship of the femoral head and acetabulum. In the absence of this relation, several characteristic anatomic abnormalities of the femur and acetabulum may develop. The term developmental dysplasia of the hip (DDH) is descriptive and appropriate. The acetabulum may be shallow, lateralized, and deficient anteriorly and superiorly. The femoral neck often has excessive anteversion with a concomitant posterior displacement of the greater trochanter and often is shortened with a coxa valga deformity. The femoral canal usually is narrow and straight. With the lack of femoral head coverage by the acetabulum and decreased surface area, the joint reactive forces and contact pressure are increased, leading to the development of arthritis. Also, the entire hemipelvis may be deficient because of disuse osteoporosis; Judet views may be helpful for further evaluation of the acetabular bone stock. Computed tomographic scans also have proven to be useful in assessing the available bone stock for optimal placement of implants, although intraoperative correlation of CT findings and recognition of appropriate cup placement may be difficult and unreliable.
In patients with dysplasia the limb often is shortened as a result of proximal migration of the femur, and associated shortening of the soft tissues occurs. The abductors lose their vertical orientation and become a transverse structure. This hampers their efficiency and also may make the arthroplasty more difficult. The femoral nerve may be more lateral and proximal than usual; therefore medial retraction should be minimized during surgery. The sciatic nerve is chronically shortened as well and is quite vulnerable if the limb is lengthened. These anatomic abnormalities should be understood and expected while proceeding with surgical planning.
INDICATIONS AND CONTRAINDICATIONS
Indications for reconstruction of an arthritic dysplastic hip are similar to those of a primary osteoarthritic process. However, these patients usually are younger. The primary indication for arthroplasty should be pain not amenable to nonoperative modalities. Any other indication may result in a dissatisfied patient. Limb-length discrepancy must be a secondary concern because this may or may not be corrected with surgical intervention. Although limp may be improved after surgery, this should not be a primary goal. Masonis et al reported worsening of the preoperative limp in 7% of their patients. Radiographic changes consistent with arthritis, such as joint space narrowing, osteophyte formation, and subchondral sclerosis, should be identified.
The patient should be actively involved in the decision-making process, especially given the difficulty of the procedure and the increased risk of poor outcomes and complications. Other options such as osteotomy and nonsurgical treatment should be considered if cartilage space remains. Osteotomy may temporarily obviate the need for arthroplasty and may provide bone stock for later reconstruction. Mechanical symptoms of catching or locking may indicate labral pathology, and with a preserved joint space magnetic resonance imaging should be considered. Positive magnetic resonance image findings with mechanical symptoms are an indication for arthroscopy.
Patients with high dislocations often present with symptoms different from the osteoarthritic hip. Buttock pain, back pain, and an abnormal gait frequently are primary symptoms, unlike the typical groin pain of osteoarthritis. Although the radiographic appearance may be dramatic in these patients, surgical decision making should not be based on the severity of the radiographs alone. Pain and dysfunction that affect daily living and quality of life remain the primary indications for surgical intervention in these patients.
Several classifications have been proposed to assess and describe the dysplastic hip. The following classifications categorize dysplasia on the basis of the femoral head position relative to the original hip center. A classification system in this setting has many goals, including facilitating communication among surgeons, aiding in the understanding and evaluation of the existing literature, and determining prognosis for differing treatments. A classification system also should help the surgeon in patient assessment and anticipation of practical issues that may arise during the procedure. These classifications are essential to comprehend and compare the reported outcomes of surgical intervention—ultimately helping the surgeon decide the optimal treatment for a particular patient.
Perhaps the most recognized system is that of Crowe et al. This classification is based on the degree of subluxation of the hip. By using a fixed point on the pelvis (the inferior teardrop) and a reproducible point on the femur (the medial head-neck junction), the degree of subluxation can be assessed. The medial head-neck junction is a reliable landmark and normally is at the level of the inferior teardrop. Dysplastic hips are then classified based on the degree of subluxation: group I, less than 50% subluxation; group II, 50% to 75% subluxation; group III, 75% to 100% subluxation; and group IV, more than 100% subluxation. For example, if the distance between the head-neck junction and the inferior teardrop is half the vertical diameter of the femoral head, this represents 50% subluxation—the upper limit of a Crowe group I. If the head is deformed, ascertaining the vertical height of the head may be impossible. In this case an assumption is made that the average vertical height of the head is one fifth that of the vertical height of the pelvis. Therefore if the distance from the medial head-neck junction to the inferior teardrop is one tenth the height of the pelvis, then the head is subluxed 50%. Figure 15-2 provides a visual representation of these measurements in a typical dysplastic hip. This classification is prognostic in that Crowe types II and III generally are more likely to have arthritic signs and symptoms requiring surgery earlier than Crowe types I and IV. In addition, the difficulty of the arthroplasty has been found to have a direct correlation with the degree of Crowe subluxation.
Hartofilakidis et al described a somewhat more practical classification system. They describe three distinct types of congenital hip disease in adults. The first type is dysplasia, with the femoral head subluxed but still contained within the original (true) acetabulum ( Fig. 15-3 ). The second type is low dislocation, in which the femoral head articulates with a false acetabulum, the inferior lip of which contacts or overlaps the superior lip of the true acetabulum ( Fig. 15-4 ). The third type is high dislocation, in which the femoral head has migrated superoposteriorly and no contact exists between the true and the false acetabulum ( Fig. 15-5 ). More severe intraoperative problems with leg-length discrepancy and bone stock should be expected when the femur has little or no contact with the true acetabulum.
A general understanding of these classification systems can help the surgeon understand and anticipate potential problems related to leg-length and cup position before the procedure begins.
As might be anticipated, virtually all complications are increased when arthroplasty is attempted in the dysplastic hip. An increased risk of nerve injury, vascular injury, infection, dislocation, and limb-length discrepancy have all been reported in the literature and should be communicated to the patient preoperatively in these cases.
The results of arthroplasty in dysplastic hips may be compromised by sciatic or femoral nerve dysfunction. The nerves to the lower extremity of a dysplastic hip grow and develop in accordance with the limb and for this reason may be chronically shortened. Therefore they are extremely intolerant to stretch and limb lengthening. In contrast, in traumatic shortening and other conditions in which the limb shortens after maturity, the nerve typically grows and develops normally along with the extremity and is relatively less susceptible when the limb is lengthened. Reported rates of nerve dysfunction after arthroplasty range from less than 1% in the general population to upwards of 5% in the patient with dysplasia. In a large series Farrell and Morrey identified the preoperative diagnosis of developmental dysplasia as a significant risk factor for the occurrence of a postoperative nerve palsy. The peroneal nerve was most commonly injured, followed by the sciatic and femoral nerves. Most of these palsies did not fully resolve. Although lengthening of the limb was associated with the development of a nerve palsy, they were unable to determine from their data a definitive maximal amount the limb can be safely lengthened. A recent report further quantified this risk. Sixty-eight hips with a high dislocation were treated with arthroplasty and a femoral shortening osteotomy. Three were complicated by a peroneal nerve palsy and one sustained a femoral nerve palsy. Woolson and Harris also documented a 5% incidence of nerve palsy after arthroplasty in dysplastic hips, with the sciatic nerve most commonly injured. An increased risk of nerve dysfunction in this patient population is a reality that must be communicated to the patient preoperatively. At the very least, if the limb is lengthened, the sciatic nerve should be identified and its tension assessed.
The use of subtrochanteric osteotomies in these complicated cases can increase the overall complication rate. Masonis et al reported a union rate of 91% after subtrochanteric osteotomy in Crowe type III and IV dysplasia, with all nonunions symptomatic and requiring revision. The possibility of nonunion and its ramifications should be understood by both the patient and surgeon before the procedure.
The rate of dislocation is increased with arthroplasty in the dysplastic hip. Although dislocation rates after total hip arthroplasty in general are sometimes difficult to ascertain, a recent meta-analysis found comparable and minimal dislocation rates with the anterolateral, lateral, and posterior approaches with soft tissue repair (0.70%, 0.43%, and 1.01%, respectively). On the other hand, Woo and Morrey report a much higher 7.5% risk of dislocation when congenital dislocation is the operative diagnosis. Dislocation rates of more than 10% have also been reported. The explanation for this increased risk is multifactorial. These patients often have had prior surgery, and the dislocation rate after arthroplasty has been established to be increased when prior surgery has been performed on the hip. However, factors related to the distorted anatomy and soft tissues have a significant influence as well. Of necessity, the size of the femoral head often is small in these patients. Berry et al found a significantly higher dislocation rate with 22-mm femoral heads relative to the 28- and 32-mm options in their large series. Dysplastic hips often require smaller implants, and a concordant increase in the dislocation rate might be expected for this reason. Offset problems also may contribute to dislocation. If the cup is left at a high hip center and medialized, anterior and/or posterior impingement is likely to occur, potentially leading to dislocation. Difficulties obtaining appropriate position and version of the femoral or acetabular components, as well as incompetence of the abductors or trochanteric nonunion, also may contribute to the increased risk of dislocation.
Finally, the risk of deep infection has been suggested to be increased in these patients. In general, prolonged operative time has been associated with deep sepsis in hip arthroplasty. Operative time in dysplastic hips often is extensive and may be an explanation for the increased risk of sepsis.
Discussion of the classifications and complications related to the dysplastic hip provide the necessary framework for evaluating the outcomes reported in the literature. Sochart and Porter have documented a marked clinical improvement after arthroplasty for dysplasia, at least in terms of pain relief, function, and range of motion. However, the need for revision is inherently higher because these patients often are young at the time of the initial surgery. Others have noted a higher rate of revision with more severely dysplastic hips. Inferior outcomes are higher in many of these cases because of the distorted anatomy and increased difficulty of the procedure.
Reports are conflicting regarding the effect of prior surgery on the outcome of total hip arthroplasty (THA) in the dysplastic hip. Previous femoral rotational osteotomy could conceivably present problems with hip arthroplasty. Increased risk of complications, such as malalignment and fracture and especially early loosening of the femoral component, have been reported with arthroplasty after femoral osteotomy. However, later reports have shown no such effect, at least with the femoral component. Boos et al compared 74 THAs carried out after previous proximal femoral osteotomy with a diagnosis-matched control group of 74 primary procedures. Their surgical exposure time was significantly longer, although early complication rates were not increased. They concluded that THA after previous osteotomy is “technically more demanding but not necessarily associated with a higher rate of complications.” On the acetabular side, the Ganz periacetabular osteotomy proposes to reorient the acetabulum—and could potentially increase the bone stock for future arthroplasty. Although this remains to be seen, at least this acetabular osteotomy does not appear to increase the technical difficulty and complications of subsequent arthroplasty.
The placement of the acetabular component is of particularly vital importance to the remainder of the arthroplasty in the dysplastic pelvis. Most problems in this area relate to bone stock and coverage of the shell. The first seminal question is “Where is the appropriate placement of the reconstructed acetabulum?” Clinical studies have reported equivocal results with leaving the hip at a high center. Russotti and Harris reviewed the results of placement of the hip at a high center and found no significant association between a high hip center and loosening. However, 16% of the acetabuli were loose at 11-year follow-up. A high hip center was deemed appropriate when host bone at the true acetabulum was not sufficient. Later reports with 14-year follow-up have shown a significant association between the high hip center and loosening and revision rates. Stans et al reported that superolateral placement of the hip center was a strong predictor of loosening at 16-year follow-up; 83% of cups placed outside the true acetabulum loosened. When lateralization occurs with the high hip center technique, increased joint reactive force with resultant loosening is likely. In addition, a high hip center can be expected to exacerbate abductor deficiency, limp, and leg-length inequality.
Although the optimal location of the hip center and placement of the acetabulum is occasionally a controversial issue, in the authors’ view the true acetabulum is almost always the optimal site for reconstruction. Recent studies have indicated a 95% survival rate with porous coated press-fit components placed in the true acetabulum at mid-term to long-term follow up. Usually the best bone stock is available in this location, and more normal hip biomechanics and abductor function are obtained.
Another difficulty in the dysplastic acetabulum is obtaining coverage of the cup. If the cup is left more than 30% uncovered, early loosening is likely. Smaller components may be necessary for coverage issues and are an obvious solution to diminished bone stock. However, a concomitant decrease occurs in the polyethylene thickness and head size with this solution. These drawbacks theoretically could adversely affect the long-term outcome of the arthroplasty. Leaving the hip at a high center is another option for coverage, as previously discussed. Medialization of the cup is an additional option for coverage. This can be accomplished by reaming or cotyloplasty—a controlled fracture of the medial wall with subsequent cementation of a cup. These techniques obviate the need for graft coverage and its inherent pitfalls. Hartofilakidis et al reported cumulative survival rates of 90% at 10 years for cotyloplasty. However, 14% of these hips had radiolucent lines greater than 1 mm around the acetabular component. Long-term follow-up is needed before definitive recommendation of this technique is appropriate. In a similar fashion Dorr et al used a medial protrusio technique with a press-fit cementless acetabular component. In their series of 24 hips none required revision for loosening at 7-year follow-up. These results are encouraging, and long-term data are anticipated. A significant concern with both these techniques is the disadvantageous loss of bone stock, which may be required for future revision. In the face of considerable bone loss trabecular metal cups have been successfully used in the revision setting. These results are encouraging for the use of trabecular metal in the face of poor host bone coverage in the dysplastic acetabulum. A final option for acetabular coverage—bone graft—is covered below.