Hip arthrodesis was a commonly performed procedure in the United States up until other forms of motion-sparing, pain-relieving reconstruction became available. However, with the advent of hip arthroplasty, each decade the number of hip arthrodeses decreases. Unfortunately for the young patient, the durability of current components of total hip arthroplasty cannot compete with patients’ lifestyle, and this often leads to revision. The success of total hip arthroplasty has improved in recent decades with quoted success of modern components in patients younger than 50 years of age to be 87% at 10 years for uncemented Harris Galante acetabular components,1 95% at 7 years for Exeter cemented femoral components, and 90% for an uncemented grit-blasted straight tapered titanium femoral stem at 20 years.² Even with exceedingly favorable results, many patients younger than 50 will need at least one revision of total hip arthroplasty in their lifetime.

Preoperative discussion in patients with end-stage arthrosis of the hip younger than 40 years of age must include discussions of the pros and cons of salvage procedures such as arthrodesis. Although surgeons’ enthusiasm for the procedure and patients’ acceptance of fusion may be low, hip arthrodesis may be an important consideration in a select patient population with end-stage arthrosis of the hip. A properly positioned, fused hip joint can offer long-term pain relief and good function.^3–6 However, as the results of total hip arthroplasty improve with further development of implants, bearing surfaces, and techniques, hip fusion may become even more limited. Discussions between patient and physician with the patient’s best interest in mind, combined with quality data, will continue to direct decision making.

Historical Perspective

In the United States in 1908, F. H. Ablee first discussed arthrodesis of the hip for advanced arthrosis in five patients.⁷ Indications for most early reports included tuberculous hips in younger patients and unilateral osteoarthritis in older patients. Arthrodesis was first used for old congenital dislocation by Heusner, by Lampugnani, and by Albert as early as 1885, according to Nové-Josserand.⁸ In the early 1900s, many methods of extra-articular arthrodesis were described.^9–11 Ghormley, in 1931, and Henderson, in 1933, described their techniques of arthrodesis, used chiefly in tuberculous hips.^12,13 Ghormley advocated a combined intra-articular and extra-articular arthrodesis and pointed out that the risk of spreading tuberculosis by opening the superior surface of the capsule was minimal. Henderson recommended clean removal of tuberculous tissue and combined intra-articular and extra-articular fusion for tuberculosis of the hip.

Trumble, in 1932,¹⁴ and Brittain, in 1941,^15,16 described techniques for ischiofemoral arthrodesis. Brittain’s method of arthrodesis involved making a subtrochanteric osteotomy with special chisels and placing a tibial autograph strut from the subtrochanteric osteotomy into the ischium. Patients were kept in plaster cast immobilization for 4 months postoperatively.

In 1938, Watson-Jones advocated internal fixation of the femoral head to the pelvis by a long Smith-Petersen nail.¹⁷ This was later refined and described in combination with iliac grafting. Watson-Jones reported in 1956 an incidence of 94% sound bone fusion in 120 patients who had arthrodesis of the hip for osteoarthritis who had been observed for a minimum of 5 years.^18,19 Immobilization in a double hip spica for at least 4 months was considered essential by these authors. Lange reported the largest series (500 patients) of the technique with 85% perfect results.²⁰

Thompson and Cholmeley, both in 1956, advocated routine subtrochanteric osteotomy for all patients having arthrodesis of the hip.^21,22 Both Thompson and Cholmeley concluded that success occurred more frequently when the grafting operation was combined with or followed by osteotomy. Thompson noted a 90% rate of union in those patients having combined hip fusion and osteotomy against a prior 26% rate of union in patients in whom fusion was attempted without osteotomy for osteoarthritis.²²

Charnley, in 1953, advocated central dislocation of the femoral head in surgical arthrodesis of the hip and subsequently reported excellent or good results in 88% of 105 patients treated by this method. Even though many of his patients did not obtain fusion, they were included in the 88% good results.²³ Medialization of the hip center lowered joint reaction forces across the hip by shortening the lever arm from the center of gravity. Schneider applied the concepts of Charnley’s arthrodesis and added the Cobra-headed plate arthrodesis.²⁴

These reports and others had relatively short follow-up and focused on fusion rates with different intra-articular and extra-articular techniques with and without internal fixation. Many reports required prolonged immobilization in plaster cast for 6 weeks to 4 months to achieve fusion.

Callaghan and Sponseller were some of the first to report the long-term outcomes of patients with arthrodesis of the hip. Sponseller cited that 78% of patients were satisfied with the arthrodesis, and all were able to work; 57% had some low back pain, and 45% had knee discomfort. Only 13% had undergone total hip arthroplasty on the arthrodesed hip.⁶ Callaghan retrospectively reviewed 28 patients with an average follow-up of 38 years after arthrodesis through various techniques. About 60% of patients had pain in the ipsilateral knee, with onset on average 23 years after arthrodesis. Back pain was similar, with an average onset of 25 years after arthrodesis. Seventy percent of patients could walk farther than 1 mile. Based on their results, investigators believed that the optimal position for fusion was in approximately 5 degrees of adduction and 35 to 40 degrees of flexion. The authors concluded that a patient with an arthrodesis of the hip could function at a high level for many years and would be able to work at most occupations. Pain in the back and knee was a common sequela, especially at long-term follow-up. However, symptoms usually are not incapacitating and generally had their onset many years after the arthrodesis.⁴

Indications/Contraindications

Preoperative Planning

An important component of the preoperative plan is a precise and careful discussion of the options and expected functional outcome of the arthrodesis. The physician-patient relationship is very important in this situation for a positive outcome. Patients should be counseled that although the motion of the hip joint will be obliterated, they will be allowed to resume all activities. The reality of the salvage situation must be conveyed to enable the patient to have realistic and appropriate expectations after the procedure.

To help define the bony anatomy and any bony deformities that may be encountered at arthrodesis, enhanced radiographic evaluation is indicated. Routine standing anteroposterior (AP) pelvis, Judet, and AP and cross-table lateral views of the femur should be obtained. These films will allow the surgeon to assess leg length discrepancy, offset, rotation, and angular deformity. Additionally, the acetabular bone stock and points of fixation can be more readily assessed with Judet films. If the pelvic anatomy or bone deficiency is more complex, preoperative three-dimensional computed tomography (CT) reconstructions of the pelvis and proximal femur may assist the surgeon in proper positioning and fixation of the arthrodesis.

Routine preoperative laboratory tests should be obtained. However, in addition, erythrocyte sedimentation rate, C-reactive protein, and complete blood count with differential are advised in patients with a history of infection. If any of these tests are abnormal, preoperative hip joint aspiration with cell count and culture will help rule out chronic infection. Patients with remote histories of inflammatory disease may have to be evaluated by a rheumatologist and may need to be reconsidered for fusion if active inflammatory disease is present. Preoperative blood donation or intraoperative blood salvage may be indicated because blood loss may necessitate transfusion.

Description of Technique(s)

Postoperative Care

Standard postoperative antibiotics are maintained for 24 hours. Drains are removed when output has declined to less than 50 mL over 8 hours. Deep venous thrombosis prophylaxis should be maintained with a comprehensive protocol as with total hip arthroplasty, including but not limited to mechanical and dynamic compression garments and chemical thromboprophylaxis. Whether casting has been chosen or not (based on the rigidity of internal fixation and surgeon preference), weight bearing should be limited to no more than 30 lb for the first 6 to 8 weeks. If not treated with a cast, the patient should not sit at any angle greater than 60 degrees so as not to stress the fixation of the fusion. AP and lateral radiographs of the pelvis are obtained at 3 months to assess the fusion, and every 6 weeks thereafter until fusion is obtained. Activity and weight bearing are progressed when bony fusion has been achieved and pain has subsided.

Results

In Callaghan’s review of 28 Iowa patients who had been fused from 1923 to 1966 with a variety of techniques, and with an average follow-up of 35 years, 22 of the 28 remained fused; in 6, the arthrodesed hip had been converted to total hip replacement. Average age at fusion was 25 years. Low back pain was present in 61%, and knee pain was present in 57%. Arthrodesis reliably relieved patients’ pain but limited lifestyle to some degree. Patients fused in abduction had more back pain than those fused in neutral or adduction (78% vs. 50% to 60%). Additionally, those fused in adduction had a lower rate of knee pain and radiographic changes in the knee than those fused in abduction (43% vs. 78%). Patients fused in a more flexed position (average 33 degrees) tended not to have back pain compared with those fused in a less flexed position, who had back pain (average 29 degrees). Most patients were unsure whether in hindsight they would have preferred a total hip arthroplasty or an arthrodesis (Table 59-1).

With more modern fusion techniques and with a group of primary fusion and re-fusion patients, Brien reviewed 16 patients who had undergone hip arthrodesis utilizing an anteriorly placed compression plate an average of 4.5 years after surgery. The rationale for the anteriorly based plate is that future conversion to total hip arthroplasty should be considered at the time of planning for hip fusion. Rates of conversion to THA range from 13% to 21%.^4,6 In this study, 63% of patients failed to unite, and 100% of those were re-fusions. Half of those with pseudarthroses were accepted by patients, and few significant restrictions in functional activities were reported.²⁵ However, in a retrospective review by Matta, at an average follow-up of 25 months, 83% achieved solid fusion by clinical and radiographic criteria. Seventy-five percent of patients in that study had no or minor restrictions when participating in their former sports activities, and half regained the ability to work at their former jobs or in new occupations.³²

Complications

Today, the major complication of arthrodesis may be patient dissatisfaction. Most patients know or have seen someone with a total hip arthroplasty and desire that level of function and hip motion. An open conversation about the procedure is very important. Proper patient education, combined with excellent surgical technique, may decrease this concern. Fusion position adjustment, based on the patient’s daily activity and interests, may enhance acceptance of an arthrodesis. Some authors have suggested a trial of spica casting before arthrodesis to “test out” the position and make modifications to the position preoperatively.

Malunion is the major complication of hip arthrodesis. Modern reports cite a high rate (≈80% to 90%) of union with modern techniques.^31,32 However, Brien found that patients with previous failed fusion or bone loss have a high risk of nonunion.²⁵ A single plate may not be adequate to neutralize forces across the hip joint, particularly in cases such as multiply operated hips, re-fusion, or avascular necrosis, when the contact surface area between the femoral head and the acetabulum is reduced. In these cases, supplemental fixation with lateral plates and additional bone graft may be indicated.²⁵