Vol 10, No 4, July/August 2002 249 Total hip arthroplasty (THA) is the standard of care for severe osteo- arthritis (OA) of the hip, especially in patients older than 50 years. In adolescents and young adults (16 to 30 years old), however, long-term durability and the prospect of mul- tiple revisions are a concern. 1 Treat- ment should relieve symptoms yet allow for as many options as possi- ble in the future. For early OA of the hip, proximal femoral and pelvic osteotomies preserve bone stock and delay or even prevent THA. 2 However, disabled patients with severe OA who desire to re- turn to an active lifestyle also may opt for a THA. 3 Hip arthrodesis is not generally perceived as a favor- able alternative by surgeons or by patients with severe OA because of the dramatic and immediate relief of pain and good functional results of joint replacement. Hip fusion is perceived as having functional out- comes inferior to those of THA. However, the long-term results of THA in this population have been disappointing, with revision rates of 33% to 45%. 4-7 In these studies, 4-7 active patients with unilateral hip disease secondary to osteonecrosis and OA had poorer results than did patients with inflammatory condi- tions. Long-term clinical data are insuf- ficient to assure prolonged sur- vivorship from newer techniques such as cementless fixation 8 or from implants utilizing alternate bear- ings. The latter includes highly cross-linked polyethylenes, 9 which may reduce the extent of peripros- thetic bone loss and osteolysis induced by wear debris. 10 The sur- geon is left with three choices: (1) delaying surgery by use of nonsur- gical modalities, such as a cane and medications; (2) performing a THA; or (3) undertaking a bone-conserving procedure such as a hip arthrodesis or osteotomy. Delaying surgery relegates the patient to a more sedentary lifestyle. A THA pro- vides reliable pain relief but puts the patient at risk for multiple revi- sion surgeries. A hip arthrodesis, when done correctly, provides pain relief, enables an active lifestyle, and may permit later conversion, if in- dicated, to a THA with minimal morbidity. Arthrodesis may be con- sidered as an option because of the long life expectancy of young pa- tients, the potential for failure of pri- mary hip replacement, and the increased risk and limited durability of revision surgery. Dr. Beaulé is Assistant Clinical Professor, Joint Replacement Institute at Orthopaedic Hospital, Los Angeles, CA. Dr. Matta is Clinical Pro- fessor, Department of Orthopaedic Surgery, University of Southern California, Los Angeles. Dr. Mast is Director, Bioregenerative Center, Northern Nevada Medical Center, Sparks, NV. Reprint requests: Dr. Beaulé, Joint Re- placement Institute at Orthopaedic Hospital, 2400 South Flower Street, Los Angeles, CA 90007. Copyright 2002 by the American Academy of Orthopaedic Surgeons. Abstract The management of young adults with severe osteoarthritis of the hip remains a problem because of the increased failure rates of total hip arthro- plasty (THA) as well as the prospect of multiple revisions in this population. Although hip arthrodesis is not perceived favorably as an alternative by most orthopaedic surgeons or patients because of the presumption of less than opti- mal functional outcomes, it is a viable technique, especially for younger patients with a recent history of local infection and/or trauma. With current internal fixation techniques, a fusion rate >80% can be achieved with maxi- mal preservation of bone stock. Proper patient selection and optimal arthrodesis position (flexion of 20° to 30°, adduction of 5°, external rotation of 5° to 10°, and limb-length discrepancy <2 cm) are essential for a success- ful, long-term result. Back and ipsilateral knee pain are the most common complaints leading to secondary conversion of a hip fusion to a THA. Symptoms improve markedly after conversion. Survivorship of the conver- sion THA is comparable to that of a primary THA when the patient is older than 50 years of age and multiple surgical procedures have been avoided. However, the procedure can be technically challenging and has a high risk of postoperative complications. J Am Acad Orthop Surg 2002;10:249-258 Hip Arthrodesis: Current Indications and Techniques Paul E. Beaulé, MD, FRCSC, Joel M. Matta, MD, and Jeffrey W. Mast, MD Results of Hip Arthrodesis The initial reports of the results of hip arthrodesis had short-term follow-ups 11 and focused mainly on fusion rates with different tech- niques (extra-articular versus intra- articular, with or without internal fixation). Indications included tuberculosis in younger patients and degenerative arthritis in older patients. Two early series that re- viewed the initial results of internal fixation reported fusion rates of 74% to 78%; surgical technique had the greatest influence on outcome. 12,13 The overall long-term results of hip arthrodesis (Table 1) depend on proper surgical technique, adequate hip positioning, minimal limb-length discrepancy, and proper patient selection. When these factors are favorable, incidence and onset of pain in adjacent joints can be signifi- cantly minimized, and most patients are able to return to an active life, including manual labor. Female patients tend to do as well as male patients, with no reported difficulties with childbirth. However, the sur- geon and the patient must be aware of the limitations imposed by the fused hip on activities of daily living as well as sexual activity, particularly when other joints become involved. The most important elements in the assessment of hip arthrodesis are functional outcome and the effect on adjacent joints (contralateral hip, ipsilateral knee, and lower back). Function and Gait Several authors 14-17 have reported satisfactory long-term function after hip arthrodesis, with most patients employed and able to walk more than 1 mile. An average of 75% of patients reported adequate pain re- lief. However, 32% experienced dif- ficulties with sexual activity, and more than 70% graded their activity as below average for their age group, although female patients younger than 18 years of age fared better than older female patients. Two smaller series of young adults with long-term follow-up 18,19 also reported that most patients had ade- quate pain relief, were able to return to work, and would consider a hip fusion again. However, these pa- tients had difficulty putting on shoes and socks 18 and had some degree of sexual impairment. 19 In a series of 40 Asian patients with hip fusion, 16 35 (87.5%) claimed that the arthrodesis limited bending during Japanese-style sitting. Other common activities affected putting on and taking off socks, standing, climbing stairs, and sexual activity. All patients returned to their previ- ous jobs, even those doing heavy labor. 20 Ahlbäck and Lindahl 21 reviewed 35 patients with a minimum 2-year follow-up. Gait was judged from an aesthetic standpoint only. Sitting was based on the patient’s activities of daily living. Alignment in the frontal plane (abduction/adduc- tion) had the greatest effect on gait because of its relationship to limb- length equality. The patients with a hip fused in 40° of flexion, a contra- lateral hip with a flexion-extension arc ≥80°, and a lumbar spine with 40° of motion exhibited the best gait. Gore et al 22 provided a more detailed analysis of gait following unilateral hip fusion after reviewing 28 men (average age, 35 years) at 6 years. A consistent gait dysrhythmia was observed secondary to a short- ened stance phase and prolonged swing phase for the fused hip com- pared with the mobile contralateral hip. Patients also exhibited a slower gait velocity because of a shortened step length. To substantially in- crease stride length, patients exhib- ited a greater than normal anterior pelvic tilt, which caused the lumbar spine to remain in varying degrees of lordosis throughout the gait cycle. The effective increase in lordosis and change in pelvic tilt resulted in the mobile hip having a greater flexion- extension excursion than normal. Also, real inequality in limb length (shortening of the fused side) as well as effective inequality (the hip po- sition in the frontal plane caused by adduction >10°) adversely affected walking performance. The greater the inequality, the more irregular the forward progression, causing greater lateral motion of the head and trunk and a tendency to walk slower. A recent study by Karol et al 23 confirmed the findings of Gore et Hip Arthrodesis Journal of the American Academy of Orthopaedic Surgeons 250 Table 1 Long-Term Results of Hip Arthrodesis Mean (yr) Pain Patient Age Low Ipsilateral Contralateral Conversion Study at Arthrodesis Follow-up Back (%) Knee (%) Hip (%) to THA (%) Sponseller et al 14 (53 hips) 14 38 57 45 17 13 Callaghan et al 15 (28 hips) 25 37 61 57 26 21 Sofue et al 16 (40 hips) 32 26 65 35 19 0 al 22 that increased motion of the lumbar spine and ipsilateral knee had a negative effect on the gait of a shortened, fused limb. Average age at the time of fusion in nine patients was 13 years. At an average follow- up of 8 years, seven patients reported earlier onset of back pain than in other series. The earlier onset of back pain was thought to be due to the younger age as well as higher activity level of the patients. Five pa- tients had good to excellent results. Contralateral Mobile Hip A review 17 of 125 patients (mean age, 52 years) with fused hips at an average 10-year follow-up revealed that those whose mobile hips showed a high probability of arthritic deteri- oration 24 (ie, asymptomatic hip dys- plasia) had an inferior functional outcome and the lowest rate of restored working capacity. These re- sults were thought to be secondary to the added stresses placed on the mobile hip during gait with a unilat- eral fusion. In another series, 16 22.5% of patients with preexisting osteoarthritic changes and/or a diagnosis of developmental dyspla- sia of the noninvolved hip had pro- gression of disease. Ipsilateral Knee and Lower Back Examining the knees of 200 pa- tients (mean age, 52 years) with unilateral hip fusion at an average follow-up of 22 years, Hauge 25 noted radiographic evidence of os- teoarthritic changes in 65% of pa- tients, with 51% exhibiting genu valgum. Most (96%) demonstrated some form of frontal or rotatory in- stability, with more than 20% of the majority complaining of knee pain or instability. No direct correlation was made between position of the fused hip and its potential effect on the ipsilateral knee. The deteriora- tion and symptoms in the ipsilateral knee were related to the rotational strain placed on the knee during the stance phase when, after the foot is placed flat on the ground, the knee compensates for the increased trans- verse pelvic rotation. These find- ings were similar to those of Spon- seller et al 14 and Callaghan et al, 15 who reported that 57% to 61% of patients had pain in the lower back and ipsilateral knee. A markedly higher incidence was noted in pa- tients with malpositioning of the fused hip (excessive abduction). In an earlier study with an average follow-up of 4.4 years, Price and Lovell 26 reported on 14 patients less than 15 years old. This group had more favorable functional results, with only one patient com- plaining of ipsilateral knee pain. THA After Hip Arthrodesis The primary indications for con- version of a hip fusion to THA are pain in the lumbar spine, ipsilateral knee, and contralateral hip. The ability to alleviate the symptoms as well as provide a functional THA at a later date is an important con- sideration when discussing the long- term outcome of a hip fusion with a patient. Hardinge et al 27 reviewed 112 hips (104 patients) converted to THA after spontaneous or surgical fusion, excluding ankylosing spon- dylitis (Table 2). After an average of 25 years of fusion, the indications for conversion were pain in the lumbar spine (71% of patients), ipsilateral knee (48.1%), contralateral mobile hip (34%), and sound, fused hips with no evidence of spinal degenera- tion (9.8%). Limb-length discrepancy >2 cm was present in 67% of pa- tients before conversion and in only 11.5% after THA. Patients whose hips were fused before puberty had less improvement in hip muscle function because of underdevelop- ment of the greater trochanter. Only 5% of patients were dissatisfied with their results. Optimum scores on hip evaluation were not achieved until 18 to 24 months after conver- sion. Strathy and Fitzgerald 28 re- ported on the long-term follow-up of 80 hips after conversion and identi- fied several risk factors for an early failure: surgical fusion (48.5% failure rate versus 5% when no previous surgery was done), more than two surgeries, and patient age of 50 years or less at the time of conversion THA. In a more detailed analysis, Kilgus et al 29 reported on 41 hips in 38 patients. Sixty-eight percent were spontaneous fusions that had re- mained fused for an average of 33 years compared with an average of 18 years for the 32% that were surgi- cally fused. A variety of total hip designs was used, including three surface arthroplasties. At the time of conversion, 68% of patients com- plained of nonradicular and activity- related back pain, and 50% com- plained of loss of function from immobility or malposition of the fused hip. Incidence of pain in the ipsilateral knee and fused hip was 42% and 16%, respectively, and 8% in both the contralateral knee and hip. The results demonstrated that a higher percentage of patients ob- tained relief of back symptoms compared with relief of pain in the ipsilateral knee, fused hip, or con- tralateral knee and hip. The range of motion was slightly less than that after primary THA. Correction of limb-length discrepancy was an important element in overall patient satisfaction. Interestingly, the UCLA hip function scores before and after THA were not significantly different, reflecting the high level of function preoperatively and the patients’ per- ception of a satisfactory result (ie, relief of back pain, correction of limb-length discrepancy). Only 33% of patients used a less restrictive device (for example, a cane instead of a crutch, or one crutch instead of two) for postoperative ambulation. Postoperatively, patients continued to improve functionally for up to 2 years. The two most important fac- Paul E. Beaulé, MD, FRCSC, et al Vol 10, No 4, July/August 2002 251 tors for postoperative abductor mus- cle strength were preoperative quali- ty and mechanical restoration of the abductor lever arm. The failure rate of the THA at a mean follow-up of 7 years (range, 2 to 16.5 years) was 8% for spontaneous fusion versus 23% in the surgically fused hips. Other risk factors for earlier failure were age less than 45 years at the time of THA and patients with two or more operations before conversion. Reikerås et al 30 reviewed 46 con- versions, with the indications for sur- gery being pain in the lower back and ipsilateral knee, as well as loss of function from immobility or malpo- sition of the fused hip. Eighty-five percent of patients were satisfied with the outcome at a mean follow- up of 8 years. Poor results (eg, pa- tients who used walking aids post- operatively or had poor abductor muscle function) were associated with a long duration of fusion and older age at the time of fusion. The common preoperative complaints of the 37 women in the group were dif- ficulties with sexual intercourse and wetting the inside of the thigh dur- ing urination, both probably second- ary to the excessive adduction of the fusion. These problems resolved after conversion. Hamadouche et al 31 reported on 45 hips after conversion, all done through a transtrochanteric approach, with a mean follow-up of 8.5 years. The indication for conversion was pain in the neighboring joints (ie, knee and lumbar spine). The only predictive factor of functional result with regard to the walking ability was the preoperative status of the gluteus muscles, which is best evalu- ated preoperatively by palpation of the contracting abductor muscles. Survivorship of hips that fused spontaneously (excluding ankylos- ing spondylitis) versus surgically was 94.6% versus 83.5% at 10 years (NS). Because abductor muscle con- traction is related to postoperative outcome, 32 Hamadouche et al felt that conversion should not be done if adequate abductor function is not present and the hip is in satisfactory position. Kreder et al 33 reviewed the com- plication rate of 40 conversions after hip arthrodesis done with a variety of surgical techniques during a 3- year period. The conversions, which represented only 0.3% of the 12,952 THAs performed, had an overall complication rate of 45% compared with 11% for primary THA. The re- vision and infection rates at 4 years were also much higher (10% for con- version versus 2% for primary THA in each category). However, the database from which this informa- tion was collected did not include the type of fusion (surgical or spon- taneous), the number of prior surgi- cal interventions, or the type of prosthesis implanted. Overall, after THA for a fused hip, patients can expect relief of pain in adjacent joints (especially the back), marked correction of limb-length discrepancies, and im- proved mobility of the hip (although not as good as with a primary THA). Gait quality, as well as postopera- tive dependence on walking aids, is related to preoperative abductor muscle function. More than half of patients require the use of walking aids after the conversion THA. Full Hip Arthrodesis Journal of the American Academy of Orthopaedic Surgeons 252 Table 2 Long-Term Results of Conversion THA After Hip Arthrodesis Patient Population Results Improvement Complications Prior Mean Age Mean Age Survivorship Good and Back Knee Disloca- Infec- Surgical at Conver- at Follow-up at Follow-up Excellent Pain Pain tions tions Study Fusion sion (yr) (yr) (%) (%) (%) (%) (%) (%) Hardinge et al 27 65 75 8.2 96.4 95 80 — 0 0 (112 hips) Strathy and 75 51 10.4 85 62.5 — — 1.2 11.2 Fitzgerald 28 (80 hips) Kilgus et al 29 32 53 7 88 76 80 66 5 7 (41 hips) Reikerås et al 30 — 58 8 85 85 — — — 0 (46 hips) Hamadouche 56 55.8 8.5 96.5 91 59 45 0 0 et al 31 (45 hips) recovery may require up to 2 years and be associated with a prolonged, intensive physical therapy program that should begin preoperatively. Survivorship of the prosthesis is comparable to that of a primary THA when multiple surgical proce- dures have not been done before conversion and if the patient is older than 50 years of age at the time of conversion 34 (Table 2). The surgery is more technically chal- lenging than a primary THA and is associated with a higher infection rate. The high incidence of infection after conversion to THA may reflect the history of sepsis or tuberculosis in many patients who have under- gone fusion. Although in some studies 27-31 no infections were active at the time of conversion, this histo- ry is likely to affect the overall infec- tion rate. Only one study 33 has attempted an assessment of the rela- tive risk of infection of conversion surgery compared with a primary THA. Although the higher rate of complications should be carefully considered and discussed with patients before proceeding with the conversion of a fused hip, conver- sion remains the preferred method to alleviate symptoms in adjacent joints, especially if the fused hip had been malpositioned. Hip Arthrodesis With Contralateral THA and Ipsilateral Knee Replacement Another approach to relieve pain is replacement of symptomatic joints if the hip is fused in the proper posi- tion. Garvin et al 35 reported on 20 patients (23 arthroplasties), with follow-up ranging from 2 to 15 years. Of the 14 replaced hips (patient age at the time of arthroplasty, 31 to 75 years), only 10 were rated as good to excellent. The other patients either needed revision surgery or still had pain in the hip or lower back. Of the nine replaced knees (patient age, 45 to 81 years), seven were available for follow-up. All required at least one postoperative manipulation, and two patients were not able to flex beyond 90°. The overall complication rate was 65%. In another series 36 of 16 total knee arthroplasties (TKAs) in pa- tients with an ipsilateral hip fusion, results in patients in whom the fused hip had been converted were comparable to those whose hips were fused in proper position. Rittmeister et al 37 reported on 18 patients with a fused hip. Eleven had conversion THA only, four had conversion THA followed by ipsi- lateral TKA, and three had TKA alone ipsilateral to the fused hip. Of the hips converted, 13 were avail- able for follow-up (average follow- up, 45 months; average age at con- version, 60.5 years). Eight patients required walking aids and had a positive Trendelenburg sign; eight had relief of back pain; and only two had relief of knee symptoms. The type of hip fusion (spontaneous versus surgical) and its duration did not affect the outcome. The three patients with fused hips who had TKAs were dissatisfied with their results; poor range of motion was the predominant problem. Of the four TKAs done after conversion of the fusion, two were rated as excel- lent and two, fair. All three studies 35-37 on the re- sults of TKA in the presence of a fused hip have reported a high com- plication rate with unpredictable outcome. Thus, the only exception to performing a TKA before con- verting the fused hip would be a patient with a satisfactorily posi- tioned hip in whom abductor mus- cle function was questionable. In these patients, the results of THA are known to be inferior, with poor gait patterns and a decreased likeli- hood of adequate knee pain relief. If the hip is fused in a poor position and the patient has significant knee pain, the conversion THA is prefer- able because of the notably inferior results of a TKA in that setting. Indications for Hip Arthrodesis The ideal candidate for hip arthrod- esis is a young adult with severe monoarticular disease, especially posttraumatic, with high activity demands and without preexisting lumbar disease or ipsilateral knee or contralateral hip arthritis. Other potential indications are a young patient in whom THA would be contraindicated or would carry a high complication rate (eg, with a his- tory of sepsis [Fig. 1] or for salvage of a multiply operated total hip). Patients with polyarticular arthritis or with bilateral developmental dysplasia of the hips in which one hip is symptomatic should not be considered because of the high like- lihood of developing contralateral hip symptoms and degenerative changes. Assessment of Hip Position The inherent inaccuracy of preopera- tive and intraoperative assessment of hip position in multiple planes has been a persistent impediment to at- tainment of optimal fusion position. Sagittal Plane (Flexion) Gore et al 22 flexed the normal hip to straighten the lumbar spine and measured the angle between the straightened lumbar spine and shaft of the femur. A comparison of clini- cal with radiologic measurements showed that radiology routinely measured more flexion, probably because of failure to flex the mobile hip sufficiently to flatten the lumbar spine completely. At the time of fusion, the amount of flexion is sim- ply measured by the angle formed by the horizontal table and femoral Paul E. Beaulé, MD, FRCSC, et al Vol 10, No 4, July/August 2002 253 shaft. Insufficient flexion will make sitting extremely difficult, and ex- cessive flexion will accentuate any shortening of the leg and put in- creased strain on the lumbar spine. Frontal Plane (Abduction and Adduction) In a neutral position, the me- chanical axis is perpendicular to a transverse axis through the pelvis (through the inferior margins of both sacroiliac joints). In this posi- tion, the femoral shaft (anatomic) axis has an average angle of 6° (5° in men, 7° in women) adduction to the vertical line representing the mechanical axis (Fig. 2). Flexion in the hip joint appears on an antero- posterior pelvic radiograph as an increased abduction angle. For increasing flexion angles, the dis- crepancy between actual and radio- graphic measurements increases and is usually 2° to 3°. According to Lindahl, 38 adduction of 3° creates a shortening of 1 cm, while abduction of 3° leads to leg lengthening of 1 cm. The apparent lengthening or shortening of the limb is purely functional and is the result of pelvic obliquity imposed in the frontal plane. These two factors must be verified when assessing hip position intraoperatively. Longitudinal Plane (Rotation) Rotation of the extremity is as- sessed visually by verifying patella and foot orientation to the level pelvis. Excessive internal rotation will tend to cause the patient to con- tinually trip over the inturned foot. Excessive external rotation of the extremity will load the knee in flex- ion across the coronal plane, pro- ducing functional problems that cause disabling symptoms in a rela- tively short time. 39 Slight exter- nal rotation is desired to facilitate putting on and taking off shoes as well as routine foot care. Limb Length If leg lengths are equal preopera- tively, the actual removal of carti- lage as well as flexion of the hip will produce an acceptable shortening of <2 cm. Compensation for a preoper- ative discrepancy of 2 to 4 cm can be achieved by abducting the leg using Lindahl’s measurements. 38 Exces- sive abduction should be avoided. For a discrepancy >4 cm, a two- stage procedure might be consid- ered because correction through limb abduction or adduction should be limited to a 2-cm difference. Variations in abduction or adduc- tion >6° have a negative effect on the overall outcome of the fusion. 21,22 The two-stage technique permits correction of limb lengths with inter- calary grafts or other lengthening techniques after the hip joint has been fused in its proper position. Recommended Optimal Positions A review of the literature (Table 3) suggests the following as optimal positions for hip arthrodesis: 20° to 30° of flexion, 5° of adduction (ana- tomic axis to horizontal line through the pelvis), and 5° to 10° of external rotation. In addition, leg shortening should be minimal (ie, limb lengths equalized within 1 to 1.5 cm). For the range of hip flexion, the activi- ties of the patient should be consid- ered. For example, if the patient spends most of the time sitting at a desk, 30° of flexion might be the op- timum, while 20° is appropriate for a manual laborer who stands most of the time. Surgical Techniques Although hip fusion may be an unfa- miliar operation to many recently trained surgeons, there are several techniques, each with specific bene- Hip Arthrodesis Journal of the American Academy of Orthopaedic Surgeons 254 Mechanical axis Transverse axis Anatomic axis Figure 2 Frontal plane alignment showing the mechanical and anatomic axes. A B Figure 1 A, Anteroposterior radiograph of a 35-year-old man 10 years after an acetabular fracture that was malreduced and complicated by a deep wound infection. B, Antero- posterior radiograph 3 years after arthrodesis with anterior plate fixation. fits and limitations. Hip arthrode- sis 11,39 began with the development of numerous techniques, many of which required lengthy postopera- tive immobilization and had high rates of failure (up to 45%). The double compression plate method of Müller and, more recently, the cobra head fixation plate 11 have provided more viable alternatives. In choos- ing a technique, the surgeon must consider later conversion to THA, missing proximal femoral bone, limb-length discrepancies, and the presence of an active infection. Regardless of technique, if active infection is present, arthrodesis should be delayed until the infection is quiescent (ie, normal laboratory test results and no active drainage). The most important factor for suc- cess of hip arthrodesis is proper positioning in the three planes. Cobra Head Plate Technique Schneider is credited with devel- oping the cobra head plate tech- nique, which is widely used 18,41,42 because of its reliable fusion rate and avoidance of postoperative cast immobilization. 11 The technique involves stripping the abductor muscles from the iliac crest to ac- commodate the cobra head of the plate together with a pelvic oste- otomy to enlarge the area of contact between femur and pelvis. Fusion rates from 94% to 100% have been reported. 40 Beauchamp et al 43 reported a modification of this technique in a series of 19 patients. By contouring the plate to fit the contour of the pelvis and proximal femur, the pelvic osteotomy was eliminated. In addition, instead of stripping the abductors, the gluteus medius was detached with a bony block and replaced with a plate and screws at the level of its original attachment. All 19 patients achieved fusion. Other surgeons have adopted the technique. 18 Stability of the implant is achieved by loading the plate in tension and the bone in compres- sion. Deficient bone stock is not uncommon, however, which can make the lateral tension band of the cobra head plate mechanically unsound (ie, with an increasing dis- tance between the plate and the loading axis, bending moments on the plate are increased). The strip- ping of the abductor muscles from the iliac wing also can negatively affect gait after THA conversion. A technique that avoids the violation of the abductor muscles would be preferable, especially if later conver- sion to THA is being considered. Anterior Plating Technique The original motivation for the anterior approach was to create a technique that provides fixation to both the pelvis and femur while sparing the hip abductor muscles. In addition, with the patient supine and the pelvis level during the surgery, positioning of the hip is facilitated. Matta et al 40 reported a fusion rate of 83% in 12 patients using an anterior plating technique through a modified Smith-Petersen approach. This technique allows placement of the plate along the pelvic brim immediately lateral to the sacroiliac joint and posterosu- perior iliac spine. With the screws in- serted in an anteroposterior direction, excellent purchase is achieved in this area of thick bone, making this technique advantageous when there is loss of acetabular or proximal femoral bone stock. The insertion of a lag screw from the trochanteric area through the supra-acetabular bone into the center of the femoral head provides additional compres- sion because of a lateral tension band effect (Fig. 3). As with other internal fixation techniques, no ex- ternal fixation (casting) is required unless the patient is expected to be noncompliant. The anterior plating technique can also be effective in the presence of loss of bone stock (Fig. 4). The patient is placed in the su- pine position on a standard fracture table or, optimally, on a Judet table. 40 On a Judet table, the hip is placed in the desired position before prepar- ing the patient. An intraoperative radiograph verifies the range of abduction-adduction. The modified Smith-Petersen approach involves elevating the abdominal muscles from the iliac crest through their fascial attachment without violating the abductor musculature. The dis- tal extension is within the tensor fascia muscular sheet, with detach- ment of both the sartorius and rec- tus femoris muscles. To expose the femur, the vastus lateralis is elevated from a lateral to medial direction to avoid denervation. With the hip joint exposed and denuded of carti- lage, the lag screw is inserted first, followed by the 12- to 14-hole low- contact broad dynamic compression plate. Viewed anteriorly, the plate Paul E. Beaulé, MD, FRCSC, et al Vol 10, No 4, July/August 2002 255 Table 3 Recommended Positions for Hip Arthrodesis Hip Lipscomb and Alhbäck and Gore Matta Karol Position McCaslin 12 Lindahl 21 et al 22 et al 40 et al 23 Flexion 15º–20º 40° 30º 15º–20º 20º–25º Add/abd 0°–10º abd 0° 0°–10º add 0°–5º add 0° External 0°–5º 5º–10° — 5º–10º Neutral rotation Add = adduction; abd = abduction has a 10° concave bend to match the internal iliac fossa, a 50° convex bend crossing the anterior acetab- ular rim, and a 35° concavity in the intertrochanteric area. Usually the plate is fixed to the pelvis first, fol- lowed by a tensioning device ap- plied to the distal end of the plate. The plate may have to be undercon- toured to avoid increasing hip flexion as the plate is being tensioned. Iliac crest bone graft from the inner table may be used if necessary. Postop- eratively, patients are usually restrict- ed to 30 pounds of weight-bearing for 8 to 10 weeks. After 12 weeks, if radiographic consolidation is present, full weight-bearing is allowed. Double-Plating Technique Double-plate fixation (Fig. 5) is particularly appropriate with diffi- cult situations such as unreduced dislocations, avascularity of bony surfaces, multiply operated hips, and poor patient compliance. However, with significant limb-length discrep- ancy (>4 cm), the arthrodesis should be individualized. In these situa- tions, a two-stage arthrodesis is often done because the capability of hip positioning to correct significant limb-length discrepancies is limited by the potential negative effect on adjacent joints (eg, increased abduc- tion associated with low back pain). The first stage is the preparation of the head and the acetabulum for fusion, usually with local fixation and an intertrochanteric osteotomy to remove the lever arm acting on the desired site of fusion. In the sec- ond stage, 6 to 8 weeks later, the intertrochanteric area is stabilized. By removing the lever arm of the femur, the fusion site may heal with greater predictability. 11 With the patient in the lateral position, a modified lateral approach is used; the gluteus medius and minimus muscles are elevated with a part of the greater trochanter. The expo- sure is continued anteriorly in the plane between the sartorius and ten- sor, with the hip flexed and exter- Hip Arthrodesis Journal of the American Academy of Orthopaedic Surgeons 256 Figure 4 A, Anteroposterior radiograph of a 16-year-old male with a combined pelvic and femoral neck fracture. He developed an intra- articular infection of his left hip 6 months after injury. Anteroposterior (B) and lateral (C) radiographs 8 months after arthrodesis. The patient maintained weight bearing as tolerated. A B C Figure 3 Anterior plating technique. Anteroposterior (A) and lateral (B) views of the pelvis with optimal position of plate and lateral lag screw. A B nally rotated. The lateral plate (broad 4.5 mm) is first applied and contoured over the trochanteric bed and placed anterior to the greater sciatic notch and along the lateral aspect of the femur. The plate is then secured proximally with a ten- sion device applied distally. After removal of the anterior-inferior iliac spine, the anterior plate (narrow 4.5 mm) is applied along the femoral shaft, and a second tensioning de- vice is applied with the plate fixed proximally. Both tensioning devices are then tightened; the plates tend to lift off the bone but are reapproxi- mated with the insertion of screws. Postoperatively, patients are limited to 30 pounds of weight-bearing for 8 to 12 weeks and allowed full weight-bearing when consolidation is evident on radiographs (Fig. 6). Summary Arthrodesis is a treatment option for young adults or adolescents with unilateral hip disease, particularly in the presence of recent infection and especially in the setting of failed pelvic or hip surgery for trauma. The ultimate goal for these patients is a return to an active lifestyle, with minimal restrictions and an accept- able rate of long-term morbidity. Arthrodesis preserves bone stock and may provide pain relief for a significant period of time. With proper patient selection and the hip fused in an optimal position, the onset of notable pain in adjacent joints can be delayed for up to 25 years. Current surgical techniques for fusion allow maximal preserva- tion of gluteus musculature, should conversion to THA eventually be considered. In patients older than 50 years of age, survivorship of the conversion THA is nearly compara- ble to that of a primary THA. Paul E. Beaulé, MD, FRCSC, et al Vol 10, No 4, July/August 2002 257 Figure 5 Double-plating technique. Drawings show optimal position of the plates in anteroposterior (A) and lateral (B) views and after reattachment of the greater trochanter (C). A B C Figure 6 A, Anteroposterior radiograph of a 16-year-old male with a painful dislocated dysplastic hip. B, Three years after fusion with the two-stage double-plating technique. A B Frontal plane 30° 60° Hip Arthrodesis Journal of the American Academy of Orthopaedic Surgeons 258 References 1. Robinson AH, Palmer CR, Villar RN: Is revision as good as primary hip replace- ment? A comparison of quality of life. J Bone Joint Surg Br 1999;81:42-45. 2. Siebenrock KA, Scholl E, Lottenbach M, Ganz R: Bernese periacetabular osteotomy. Clin Orthop 1999;363:9-20. 3. Wright JG, Rudicel S, Feinstein AR: Ask patients what they want: Evalua- tion of individual complaints before total hip replacement. J Bone Joint Surg Br 1994;76:229-234. 4. Dorr LD, Luckett M, Conaty JP: Total hip arthroplasties in patients younger than 45 years: A nine- to ten-year follow-up study. Clin Orthop 1990;260: 215-219. 5. Ortiguera CJ, Pulliam IT, Cabanela ME: Total hip arthroplasty for osteo- necrosis: Matched-pair analysis of 188 hips with long-term follow-up. J Arthro- plasty 1999;14:21-28. 6. Sochart DH, Porter ML: Long-term re- sults of cemented Charnley low-friction arthroplasty in patients aged less than 30 years. J Arthroplasty 1998;13:123-131. 7. Torchia ME, Klassen RA, Bianco AJ: Total hip arthroplasty with cement in patients less than twenty years old: Long-term results. J Bone Joint Surg Am 1996;78:995-1003. 8. McLaughlin JR, Lee KR: Total hip arthroplasty in young patients: 8- to 13-year results using an uncemented stem. Clin Orthop 2000;373:153-163. 9. McKellop H, Shen F-W, Lu B, Camp- bell P, Salovey R: Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res 1999;17:157-167. 10. Schmalzried TP, Callaghan JJ: Wear in total hip and knee replacements. J Bone Joint Surg Am 1999;81:115-136. 11. Liechti R (ed): Hip Arthrodesis and Associated Problems. Berlin, Germany: Springer-Verlag, 1978. 12. Lipscomb PR, McCaslin FE Jr: Arthrod- esis of the hip: Review of 371 cases. J Bone Joint Surg Am 1961;43:923-938. 13. Stewart MJ, Coker TP Jr: Arthrodesis of the hip: A review of 109 patients. Clin Orthop 1969;62:136-150. 14. Sponseller PD, McBeath AA, Perpich M: Hip arthrodesis in young patients: A long-term follow-up study. J Bone Joint Surg Am 1984;66:853-859. 15. Callaghan JJ, Brand RA, Pedersen DR: Hip arthrodesis: A long-term follow- up. J Bone Joint Surg Am 1985;67: 1328-1335. 16. Sofue M, Kono S, Kawaji W, Homma M: Long term results of arthrodesis for severe osteoarthritis of the hip in young adults. Int Orthop 1989;13:129-133. 17. Gudmundsson G: Function following arthrodesis for coxarthrosis with spe- cial reference to the mobile hip. Acta Orthop Scand Suppl 1972;141:1-79. 18. Roberts CS, Fetto JF: Functional out- come of hip fusion in the young pa- tient: Follow-up study of 10 patients. J Arthroplasty 1990;5:89-96. 19. Barnhardt T, Stiehl JB: Hip fusion in young adults. Orthopedics 1996;19: 303-306. 20. Sofue M, Kono S, Kawaji W, Homma M: Long term results of arthrodesis for severe osteoarthritis of the hip in young adults. Int Orthop 1989;13: 129-133. 21. Ahlbäck SO, Lindahl O: Hip arthrode- sis: The connection between function and position. Acta Orthop Scand 1966;37:77-87. 22. Gore DR, Murray MP, Sepic SB, Gardner GM: Walking patterns of men with unilateral surgical hip fu- sion. J Bone Joint Surg Am 1975;57: 759-765. 23. Karol LA, Halliday SE, Gourineni P: Gait and function after intra-articular arthrodesis of the hip in adolescents. J Bone Joint Surg Am 2000;82:561-569. 24. Murphy SB, Ganz R, Muller 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. 25. Hauge MF: The knee in patients with hip joint ankylosis: Clinical survey and bio-mechanical aspects. Acta Orthop Scand 1973;44:485-495. 26. Price CT, Lovell WW: Thompson arthrodesis of the hip in children. J Bone Joint Surg Am 1980;62:1118-1123. 27. Hardinge K, Murphy JC, Frenyo S: Conversion of hip fusion to Charnley low-friction arthroplasty. Clin Orthop 1986;211:173-179. 28. Strathy GM, Fitzgerald RH Jr: Total hip arthroplasty in the ankylosed hip: A ten-year follow-up. J Bone Joint Surg Am 1988;70:963-966. 29. Kilgus DJ, Amstutz HC, Wolgin MA, Dorey FJ: Joint replacement for anky- losed hips. J Bone Joint Surg Am 1990; 72:45-54. 30. Reikerås O, Bjerkreim I, Gundersson R: Total hip arthroplasty for arthrodesed hips: 5- to 13-year results. J Arthroplasty 1995;10:529-531. 31. Hamadouche M, Kerboull L, Meunier A, Courpied JP, Kerboull M: Total hip arthroplasty for the treatment of anky- losed hips: A five to twenty-one-year follow-up. J Bone Joint Surg Am 2001; 83:992-998. 32. Amstutz HC, Sakai DN: Total joint replacement for ankylosed hips: Indi- cations, technique, and preliminary re- sults. J Bone Joint Surg Am 1975;57:619-625. 33. Kreder HJ, Williams JI, Jaglal S, Axcell T, Stephen D: A population study in the province of Ontario of the compli- cations after conversion of hip or knee arthrodesis to total joint replacement. Can J Surg 1999;42:433-439. 34. Finerman GAM, Dorey FJ, Grigoris P, McKellop HA (eds): Total Hip Arthro- plasty Outcomes. New York, NY: Churchill Livingstone, 1997. 35. Garvin KL, Pellicci PM, Windsor RE, Conrad EU, Insall JN, Salvati EA: Contralateral total hip arthroplasty or ipsilateral total knee arthroplasty in patients who have a long-standing fusion of the hip. J Bone Joint Surg Am 1989;71:1355-1362. 36. Romness DW, Morrey BF: Total knee arthroplasty in patients with prior ipsilateral hip fusion. J Arthroplasty 1992;7:63-70. 37. Rittmeister M, Starker M, Zichner L: Hip and knee replacement after long- standing hip arthrodesis. Clin Orthop 2000;371:136-145. 38. Lindahl O: Determination of hip adduction, especially in arthrodesis. Acta Orthop Scand 1965;36:280-293. 39. Fiddian NJ: Arthrodesis of the hip, in Reynolds D, Freeman M (eds): Osteo- arthritis in the Young Adult Hip: Options for Surgical Management. Edinburgh, UK: Churchill Livingstone, 1989, pp 155-168. 40. Matta JM, Siebenrock KA, Gautier E, Mehne D, Ganz R: Hip fusion through an anterior approach with the use of a ventral plate. Clin Orthop 1997;337: 129-139. 41. Barmada R, Abraham E, Ray RD: Hip fusion utilizing the cobra head plate. J Bone Joint Surg Am 1976;58:541-544. 42. McGraw RW, Duncan CP, Beauchamp CP: Arthrodesis: The Vancouver Technique, in Sledge CB (ed): Master Techniques in Orthopaedic Surgery: The Hip. Philadelphia, PA: Lippincott- Raven, 1998, pp 197-207. 43. Beauchamp CP, Duncan CP, McGraw RW: Abstract: Don’t throw away the reamers: A new technique of hip arthrod- esis. J Bone Joint Surg Br 1985;67:330.