Abstract
Background: Revision of an acetabular component in a patient who has severe periacetabular bone loss is a complex problem, particularly when there is not enough bone stock to allow placement of an acetabular component near the normal anatomical hip center. A valuable option for revision in such a situation is placement of a hemispherical shell, fixed with screws and without cement, against the superior margin of the acetabular defect. The resulting hip center is more proximal than that seen following a typical primary total hip replacement.Methods: Forty-six hips in forty-four patients were treated consecutively, between July 1984 and February 1988, with a revision in which a hemispherical acetabular component was fixed with screws and without cement. All shells but one were placed with a so-called line-to-line fit. The procedures resulted in a so-called high hip center—that is, the center of rotation of the revised hip was located at least thirty-five millimeters proximal to the interteardrop line. The mean age of the patients at the time of the index procedure was fifty-two years (range, twenty-five to eighty-one years). The most common diagnosis for which the original arthroplasty was performed was osteoarthritis secondary to congenital hip dysplasia or dislocation (twenty-two hips). Thirty-four hips had had a high hip center before the index revision, and most patients had had a substantial limb-length discrepancy, with a mean of 1.6 centimeters of shortening on the side of the operation. In thirty-three hips, the femoral component was replaced as well, with a long-neck or calcar-replacement stem used when necessary to maintain or increase the length of the limb.Results: Six patients (six hips) died before the minimum eight-year follow-up interval; none had had another revision or loosening of the revised acetabular component. Of the remaining patients, four (four hips) had the implant removed. One of them had a resection arthroplasty and one of them had a hip disarticulation because of infection after a subsequent femoral reoperation. Another had a hip disarticulation because of late infection. The fourth implant was removed because it had displaced into the pelvis at approximately six years; this was the only reoperation for aseptic loosening in the series. The remaining thirty-six hips (thirty-four patients) were followed for a mean of 10.4 years (range, 8.5 to 12.7 years). One acetabular component migrated medially and was scheduled for revision. No other acetabular component was loose or had been revised. The mean Harris hip score was 81 points (range, 56 to 100 points) at the time of the most recent follow-up. Despite the use of a high hip center, the prevalence of a positive Trendelenburg sign was reduced from 98 percent (forty-five of forty-six hips) preoperatively to 44 percent (sixteen of thirty-six hips) at the time of the most recent follow-up. The short limbs were lengthened a mean of seven millimeters (range, five millimeters of shortening to forty millimeters of lengthening).Conclusions: In this study of acetabular revisions with use of a high hip center in patients who had major periacetabular bone loss, mechanical failure occurred in 4 percent (two) of the forty-six hips in the entire series and in 6 percent (two) of the thirty-six hips in patients who were alive and still had the implant in place after a mean of 10.4 years of follow-up. The use of a high hip center did not adversely affect function of the abductor muscles, and the mean limb-length discrepancy was reduced by the femoral reconstruction.
Revision of an acetabular component when there is loss of bone stock poses a distinct set of problems for the hip surgeon. The approach to each patient depends on the severity and location of the acetabular bone loss. Contained acetabular defects can generally be treated by grafting of the defects and insertion of a hemispherical acetabular component with screws and without cement. Good-to-excellent intermediate-term results can be expected with use of this technique11,20,38,42. This approach can be successful even in patients who have more severe bone loss (such as the absence of the medial wall in combination with defects in the anterior or posterior column), as was demonstrated in one study at a mean of seven years after insertion of large-diameter acetabular components without cement6. Frequently, however, the resulting hip center is more proximal than that seen following a primary hip replacement.
If there is insufficient viable osseous support at the normal hip center, as is often the case when a cemented acetabular component has migrated proximally or when osteolysis has destroyed the osseous support superiorly (Figs. 1-A, 1-B, and 1-C), revision without cement at the normal hip center may not be possible. Abandoning the acetabular component altogether and using a large-head bipolar prosthesis has not proved to be a reliable approach in this situation5,25,26,28,29. One viable option is augmentation of the deficient acetabulum with a structural graft or massive amounts of particulate graft, allowing fixation of an acetabular component with cement close to the normal hip center. However, our long-term experience with bulk structural grafts has shown that the rate of failure increases with time24, with two-thirds of the acetabular components becoming loose by sixteen years37. Because of the tendency of bulk grafts to collapse or resorb, or both, over time, some authors have advocated supporting the acetabular component with an acetabular protrusio cage when there is inadequate acetabular bone stock2,15,28,30. Although encouraging short-term results have been reported with this technique, the long-term outcomes remain unknown2,12,32,40,44.
Another option is to fill the acetabular void superiorly with metal, in the form of an oblong, bilobed acetabular component17. The superior of the two joined hemispheres is in contact with the host bone above, allowing establishment of a normal hip center. In many patients, it is necessary to remove additional acetabular bone in order to accommodate the shape of the oblong implant. Furthermore, these components are more costly than standard, hemispherical shells.
A valuable alternative to augmentation of the deficient acetabulum is placement of a hemispherical shell, fixed with screws and without cement, against the superior margin of the acetabular defect. The resulting hip center usually is more proximal than that seen following a typical primary total hip replacement, although often it is unchanged or even moved distally compared with that before the revision.
We arbitrarily defined a high hip center as a center of rotation of the femoral head located at least thirty-five millimeters proximal to the interteardrop line—that is, more than twice the distance proximal to the teardrop than is present in a normal hip33,36. Correction of limb-length inequality frequently requires a long-neck or calcar-replacement femoral component. In order to restore the tension of the abductor muscles, the greater trochanter often must be advanced. Furthermore, because of the proximal location of the reconstructed hip, impingement may occur between the femoral component or the femur and the pelvis. This may necessitate removal of portions of the anterior column, anterior superior iliac spine, ischium, or greater trochanter in some patients. We report the results of acetabular revision resulting in a hip center at least thirty-five millimeters proximal to the interteardrop line in a consecutive series of forty-six hips in forty-four patients.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was the William H. Harris Foundation.
†Fremont Orthopaedic Medical Group, 38690 Stivers Street, Fremont, California 94536.
‡Orthopaedic Biomechanics Laboratory, GrJ 1126, Massachusetts General Hospital, 55 Fruit Street, Boston, Massachusetts 02114.
Study Group
Between July 1984 and February 1988, ninety-eight revisions of the acetabular component were performed at our institution by the senior one of us (W. H. H.). In forty-six hips (forty-four patients), the postoperative hip center was located at least thirty-five millimeters proximal to the interteardrop line because of loss of acetabular bone. These patients form the basis of the present study.
There were thirty women and fourteen men. The mean age at the time of the index procedure was fifty-two years (range, twenty-five to eighty-one years). A mean of 2.2 (range, one to eight) previous hip operations had been performed. Four patients had had a previous acetabular revision, and three patients had had a previous resection arthroplasty because of infection. The diagnosis at the time of the original arthroplasty was osteoarthritis secondary to congenital dysplasia or dislocation in twenty-two hips, osteonecrosis in six, osteoarthritis without sufficient information to determine the cause in five, osteoarthritis secondary to developmental so-called pistol-grip deformity in three, posttraumatic osteoarthritis in three, and Legg-Calvé-Perthes disease, rheumatoid arthritis, juvenile rheumatoid arthritis, pigmented villonodular synovitis, childhood infection, tuberculous arthropathy, and ankylosing spondylitis in one hip each. The mean weight of the patients was sixty-five kilograms (range, forty-two to 109 kilograms).
Forty-five hips were associated with a positive Trendelenburg sign before the index operation. The mean limb-length discrepancy was 1.7 centimeters of shortening on the involved side (range, six centimeters of shortening to two centimeters of greater length). Two of the involved limbs had been longer than the contralateral limb before the index operation. The mean preoperative Harris hip score was 52 points (range, 30 to 75 points).
All of the procedures were carried out in a clean-air laminar-flow room with body-exhaust suits worn by the operating team. Trochanteric osteotomy was used for wide exposure of the acetabular dome in forty-four hips. In most patients, release of the iliopsoas tendon was necessary as well, in order to mobilize the femur adequately. Thirty-three osteotomies were of the horizontal type22, and six were vertical because of a previous trochanteric advancement. Five standard trochanteric osteotomies were done with the lateral origin osteotomy beginning just distal to the vastus tubercle; all five were done in hips in which a proximal circumferential femoral allograft was needed. In thirty-three hips, the trochanteric fragment was advanced and secured to the lateral femoral cortex.
The acetabular bone stock was inspected, and the approach for the reconstruction was selected. If there was adequate osseous support to allow placement of a large-diameter acetabular component with the resulting hip center close to the normal level, this technique was preferred. If, however, there was not enough bone stock for use of a large component, the approach that resulted in a high hip center, generally with use of a smaller component placed in the superior extent of the acetabular cavity, was selected. The remaining acetabulum was shaped into a hemisphere, forming an apex that was more proximal than the lateral edge of the acetabulum. Contained defects were packed with autologous bone or particulate allograft. After the hemisphere has been formed, there must be sufficient bone stock to resist strong manual pressure placed on the trial acetabular component in the direction of the resultant hip force. There must also be sufficient stability to resist medial or posterior displacement of the trial component when a force is directed in those directions.
A Harris-Galante-I acetabular component (Zimmer, Warsaw, Indiana) was used in forty-two hips, and a Harris-Galante-II component (Zimmer) was placed in the remaining four hips. The principal changes incorporated into the Harris-Galante-II component include a larger screw diameter (6.5 millimeters compared with 5.1 millimeters for the Harris-Galante-I component), a thicker shell, more tines for fixation of the liner, and more screw-holes. The mean outer diameter of the acetabular component was fifty-five millimeters (range, forty to seventy-four millimeters). Although six of the shells were considered jumbo components (sixty-six millimeters or more in diameter), these reconstructions still resulted in a high hip center. Except in one hip, the nominal diameter of the shell matched that of the last reamer used (that is, a so-called line-to-line fit). Screws (mean, 3.4; range, two to nine) were used in all hips. The estimated contact with host bone averaged 79 percent (range, 50 to 90 percent).
The severity of the bone loss observed at the time of the revision was classified according to a previously described scheme42. Stage-I acetabula had small defects that could be treated with additional reaming of the acetabulum, eliminating the need for bone-grafting. Stage-II defects were those that, despite reaming, needed treatment with particulate bone graft or fragments of bone from the reaming, or both, but did not compromise the integrity of the acetabular cavity. Stage-III defects were larger, often full-thickness, and mandated more substantial grafts to reconstruct the acetabulum. These defects were further subdivided into stages IIIA and IIIB. In stage IIIA, bone graft was generally used to fill large defects present after the removal of cement from the previous keyholes as well as isolated defects in the medial wall as large as ten millimeters in diameter. Stage-IIIB acetabula had more severely compromised bone, such as larger defects in the medial wall or the anterior or posterior column, or a combination of these defects. Stage-IV defects were rare and consisted of pelvic discontinuity or such severe destruction that reconstruction was very difficult and necessitated use of a bulk structural allograft.
The acetabular bone stock was classified as stage I in two hips, stage II in five, stage IIIA in nineteen, stage IIIB in eighteen, and stage IV in two. Most of the patients who had stage-IIIA or more severe defects had defects in the medial wall (as large as thirty-six square centimeters), often in combination with deficiency of the anterior column or enlarged anchoring holes for cement, or both. Forty-two hips were treated with particulate graft, usually obtained from the acetabular reaming but at times in combination with particulate iliac crest graft or morseled allograft. Four hips did not need bone-grafting. An allograft femoral head was used for structural support in one stage-IV hip.
Resection of portions of the ischium, anterior column, or greater trochanter was necessary in twenty-two hips in order to avoid impingement between the femur, greater trochanter, or femoral component and the pelvis at the extremes of motion. In seven of these hips, the location of the reconstruction necessitated excision of the anterior inferior iliac spine and release of the direct head of the rectus femoris. The site of impingement was most commonly posterior with the hip in external rotation; this necessitated an isolated partial ischiectomy in fifteen hips.
The femoral component was revised in thirty-three of the forty-six hips. Twenty-six femoral components were revised because of aseptic loosening; two, because of femoral lysis; and one each, because of fracture of the stem, fracture of the femur, and retroversion of the component. Two femoral revisions were reoperations after resection arthroplasty. No well fixed femoral component was exchanged because of limb-length discrepancy. Nineteen stems were fixed with cement, and fourteen were inserted without cement. Three stems were calcar-replacement components, twenty had a long neck, nine had a medium neck, and one had a short neck. The diameter of the femoral head was thirty-two millimeters in three hips, twenty-eight millimeters in ten, twenty-six millimeters in eighteen, and twenty-two millimeters in two.
Follow-up
All patients were followed prospectively both clinically and radiographically. Six patients (six hips) died before the minimum follow-up interval of eight years, and the last known status of the prosthesis before death was reviewed. Four additional patients (four hips) had the implant removed. Three were removed because of late infection, as will be described. All thirty-six hips in which the implant had been retained were followed clinically with the use of detailed questionnaires and determination of the Harris hip score16. In addition, we performed a physical examination on twenty-one hips, and other orthopaedic surgeons examined four.
An anteroposterior radiograph of the hips and pelvis, a true lateral radiograph of the hip, and a frog-leg lateral radiograph of the femur were made at the time of the most recent follow-up for thirty-four of the thirty-six hips. Right and left 45-degree oblique radiographs of the pelvis were also made for twenty of these hips. The two patients (two hips) without current radiographs had completed questionnaires and had been interviewed by telephone.
All preoperative, postoperative, and most recent radiographs were evaluated by an orthopaedic surgeon other than the senior one of us. Gaps were defined as sharp radiolucent lines between the shell and the pelvis seen on the initial postoperative radiographs, and they were believed to represent incomplete seating of the implant or defects remaining from the previous reconstruction34. Radiolucent lines were defined as lines at the bone-shell interface that had not been seen on the initial postoperative radiographs or as gaps that had increased in size after the two-year radiographic evaluation34. Any radiolucent line greater than three millimeters in width was classified arbitrarily as osteolysis. Periacetabular radiolucency was classified according to its location with use of the method of DeLee and Charnley8.
The position of the hip center was determined in reference to a line drawn between the inferior margins of the teardrops13,33. A change in the position of the acetabular component of four millimeters or more was defined as migration. Before the index operation, thirty hips had acetabular osteolysis in at least one zone and nineteen had migration of the acetabular component of six to twenty-two millimeters. At that time, the mean location of the hip center (and standard deviation) was 43.5 ± 11.4 millimeters (range, twenty to seventy-seven millimeters) proximal to the interteardrop line and 35.5 ± 9.8 millimeters (range, nineteen to sixty-two millimeters) lateral to the inferior margin of the ipsilateral teardrop. Thirty-four hips were classified as having a high hip center before the index revision.
Polyethylene wear was measured with a digital micrometer with the technique of Livermore et al.21, modified for use for acetabular components fixed without cement as described previously7. The initial postoperative anteroposterior radiograph was reviewed to confirm that the femoral head was concentrically positioned in the acetabular component. On the most recent anteroposterior radiograph, the center of the femoral head and the center of the acetabular shell were identified with use of specially prepared concentric circle guides and were marked with a fine pinhole. The deviation of the center of the femoral head from the center of the acetabular shell indicated the vector of maximum wear in the plane of the radiograph. The change in the thickness of the polyethylene-shell composite on the most recent radiograph compared with that seen on the initial postoperative radiograph was measured along this vector and was corrected for magnification with use of the diameter of the femoral head measured on each radiograph. If the center of the femoral head or that of the shell could not be clearly identified because the margins were obscured, wear was not measured.
The Student t test was used to compare the mean polyethylene wear rates among various groups of patients. A probability level of p < 0.05 was used to determine significance.
No patient was lost to follow-up. Of the six patients (six hips) who died before the minimum follow-up interval of eight years, none had had loosening of the acetabular component or a reoperation. Five had a well functioning hip at the time of death. The sixth patient had extensive femoral osteolysis and had sustained a fracture through the lytic area in the femur at 7.1 years postoperatively. One patient had had recurrent dislocations, which had been treated successfully with an abduction brace before the patient died at 4.3 years. Another patient had asymptomatic femoral osteolysis at the time of death at 5.2 years.
Of the remaining thirty-eight patients (forty hips), one, a seventy-eight-year-old obese woman, had a resection arthroplasty after the acetabular component broke and displaced into the pelvis. The acetabular bone stock had been graded as stage IIIB at the time of the index revision, and extensive particulate graft had been used. The forty-eight-millimeter-diameter acetabular component that was used had little intrinsic stability, and four screws were needed for fixation. At 5.8 years, the acetabular component protruded through the medial wall and the patient was not considered to be a candidate for reconstruction. The resection arthroplasty was the only reoperation performed because of aseptic loosening in this series.
Of the three hips that became infected, two did so after a subsequent femoral reoperation. One of these hips, in a forty-eight-year-old woman, became infected with both Staphylococcus aureus and Streptococcus milleri at 8.8 years after the index procedure and nine months after a repeat revision of the femoral component because of osteolysis. A resection arthroplasty was performed. The second infection developed in a twenty-seven-year-old woman with juvenile rheumatoid arthritis who had been managed with a proximal circumferential femoral allograft at the time of the index revision. The hip became acutely infected with Staphylococcus aureus after a subsequent procedure to place bone graft at the site of a delayed union of the allograft-host junction. Ultimately, a hip disarticulation was performed because of extensive infection in the massive femoral allograft. The third infection developed in a patient who had a complex history, including tuberculous arthropathy, a cup arthroplasty, and a previous infection treated with open drainage at the site of a total hip arthroplasty. Seven years after the index revision, when the patient was eighty-eight years old, a deep infection developed. This was treated successfully for 1.5 years with débridement and suppressive antibiotics at an outside institution, but she had a hip disarticulation following a stroke with severe ipsilateral paralysis associated with recurrent dislocations, in addition to the persistent infection. Thus, the overall rate of infection was 7 percent (three of forty-six). However, if the patients in whom the infection developed after a subsequent femoral reoperation are excluded, the infection rate was 2 percent (one of forty-four).
The remaining thirty-four patients (thirty-six hips) were followed for a mean of 10.4 years (range, 8.5 to 12.7 years). The mean Harris hip score for these thirty-six hips was 81 points (range, 56 to 100 points) at the time of follow-up. Nine hips were considered to have had an excellent result (a hip score of 90 points or more); twelve, a good result (a hip score of 80 to 89 points); eight, a fair result (a hip score of 70 to 79 points); and seven, a poor result (a hip score of less than 70 points). Eleven hips were not painful, fifteen occasionally caused slight pain, eight were mildly painful, and two were moderately painful.
Of the seven hips that had a poor result, five had a hip score of at least 65 points and two had a score of 56 points. One hip was not painful, one occasionally caused slight pain, four were mildly painful, and one was moderately painful. Four of the seven poor results were due to considerable debilitation from factors unrelated to the hip, including polyarticular rheumatoid arthritis, severe poliomyelitis with an arthrodesis of the contralateral hip (Figs. 2-A, 2-B, 2-C, 2-D and 2-E), advanced Parkinson disease, and moderate arthritis involving the ipsilateral knee in one patient each. A fifth patient, who had a hip score of 65 points, had sustained a fracture of the greater trochanter, which migrated, producing a severe limp. She had only slight, occasional pain in the hip. The sixth patient (hip score, 66 points) had a loose acetabular component (as will be described) and was scheduled for revision as of the latest follow-up evaluation. The seventh patient (hip score, 56 points) had a loose femoral component, which had been inserted before the index acetabular revision.
Of the thirty-six acetabular components that remained in place in living patients, one had migrated at ten years and was scheduled for revision (as just mentioned). This patient was forty-eight years old at the time of the index revision. The primary total hip replacement, performed with cement, had been done at another institution because of congenital hip dislocation and had resulted in a high hip center (forty-five millimeters proximal to the interteardrop line). At the index operation, the acetabular bone stock was graded as stage IIIA, with a one-centimeter-diameter defect in the medial wall and extensive osteolysis in zone III8. Particulate autogenous graft from the iliac crest and graft obtained from the reaming was used, and because the acetabulum was not suitable for reconstruction with a jumbo component we placed a forty-eight-millimeter-diameter acetabular shell against the superior portion of the acetabular recess and fixed it with three screws. The new hip center was located fifty-two millimeters proximal to the interteardrop line and, with the small-diameter shell, no inferior support by the pubis was possible. Gaps were seen in all three zones on the initial postoperative anteroposterior radiograph and in two zones on both of the oblique radiographs. The fixation failed, and the component eventually migrated.
No other acetabular component was loose at the time of the most recent follow-up; thus, the rate of mechanical loosening was 6 percent (two of thirty-six). Ten hips had radiolucency in all three zones8 of the acetabular implant-bone interface as seen on at least one of the three pelvic radiographs. In two of these hips, the radiolucent line appeared to be continuous. One of the two had a 0.5 to one-millimeter-wide line, and the other had a similar line as well as a four-millimeter-wide line in zone III on the obturator oblique radiograph. We found no association between the presence and width of radiolucent lines and the appearance of gaps on the initial postoperative radiographs or the clinical status of the hip. We also detected no association between radiolucent lines at the interface and the stage of the acetabular bone stock or the estimated original contact between the acetabular shell and the host bone.
Two hips showed evidence of acetabular osteolysis. Both had osteolysis around a single acetabular screw, and one also had it in zone III of the implant-bone interface. The patient who had the osteolysis only around a screw is the same patient who had a loose acetabular component that was scheduled for revision. The other patient is the woman who had the four-millimeter-wide radiolucent line in zone III. At the time of the latest follow-up, she had an active lifestyle. She had a titanium femoral head and a rate of linear wear of the polyethylene of 0.22 millimeter per year. The osteolysis was first identified on the most recent radiographs. The hip score was 88 points.
Of the thirty-three femoral components that had been placed at the time of the index acetabular revision, one component with a noncircumferential porous coating that had been inserted without cement was subsequently revised at eight years postoperatively because of femoral osteolysis without loosening. The hip became infected nine months after this femoral revision, as was described, and the implant had to be removed. Two additional femoral components that had been inserted without cement were revised because of osteolysis; one was revised at 9.1 years and the other, at 5.3 years. Two femoral components were seen to be loose radiographically at the time of the latest follow-up, and a third stem was seen to be loose before a patient's death.
Dislocation occurred in five (11 percent) of the forty-six hips, with recurrent dislocations occurring in three of them. One of these three hips was in the woman who had a deep infection followed by a stroke, as was described. The other two hips with recurrent dislocations were managed with an abduction brace, and there were no subsequent dislocations. One of the patients with recurrent dislocations had had one centimeter of shortening of the limb without advancement of the greater trochanter; this patient subsequently sustained a fracture of the greater trochanter.
Five (11 percent) of forty-four greater trochanteric fragments failed to unite, and three of them migrated proximally. A previous trochanteric osteotomy had been performed in four of the five hips, and a previous nonunion had been present in three. None of these hips needed a reoperation. Additional complications included one femoral and one sciatic nerve palsy, both of which resolved, and one instance of arterial bleeding, which was controlled by arterial embolization39.
Three additional patients had a reoperation after the index revision. One patient had bone-grafting on the femoral side, at three weeks after the index procedure, because of an unrecognized femoral perforation. The second patient had a revision of a loose femoral component that had been inserted before the index revision. The third patient sustained a periprosthetic fracture of the femur at 7.1 years postoperatively. The fracture was adjacent to a well fixed cemented femoral component that had been placed before the index revision; it was treated with open reduction and internal fixation with a plate, screws, and cerclage wires.
Polyethylene wear was measurable on the most recent anteroposterior radiographs of thirty-three hips, at a mean of 10.0 years postoperatively. Linear wear averaged 0.17 millimeter per year, with a relatively wide range (0.03 to 0.47 millimeter per year). The patient with the most linear wear had a Trapezoidal-28 femoral component (Zimmer) that had been inserted before the index revision. In our small series, we detected no association between the linear wear and the age, weight, or gender of the patient; the duration for which the prosthesis had been in place; the type of fixation of the stem; the size of the femoral head; the abduction of the cup; or the presence of osteolysis. The hips in which the femoral component had not been revised had a slightly higher mean rate of linear wear (0.21 ± 0.12 millimeter per year) than the hips in which both components had been revised (0.16 ± 0.10 millimeter per year); however, with the numbers available, this difference was not found to be significant.
Although on the average the location of the hip center at the time of follow-up (44.5 ± 9.5 millimeters proximal to the interteardrop line) was virtually unchanged from the preoperative location (43 ± 1.4 millimeters proximal to the interteardrop line), the location changed in most patients as a result of the operation. In twenty hips, the hip center was moved a mean of 7.4 millimeters (range, one to twenty-three millimeters) proximally and a mean of 2.0 millimeters medially. In nineteen hips, the hip center was moved a mean of 6.7 millimeters (range, one to twenty millimeters) distally and a mean of 2.0 millimeters laterally. In five of the hips in which the hip center was moved distally, a jumbo acetabular component had been used.
Forty-five hips had been associated with a positive Trendelenburg sign preoperatively. At the time of the most recent follow-up, fifteen of these hips continued to be associated with a positive Trendelenburg sign and the hip that had been associated with a negative Trendelenburg sign preoperatively demonstrated a positive Trendelenburg sign. Twenty-seven patients (twenty-nine hips, including the sixteen just mentioned) reported some degree of limp at the time of the most recent follow-up.
Limb-length discrepancies were generally decreased postoperatively. Before the index revision, the mean limb-length discrepancy was 1.6 centimeters of shortening (range, six centimeters of shortening to two centimeters of greater length) on the side of the operation. Postoperatively, the mean discrepancy was 0.9 centimeter of shortening (range, five centimeters of shortening to two centimeters of greater length) on the side of the operation. The short limbs were lengthened a mean of seven millimeters (range, five millimeters of shortening to forty millimeters of lengthening). Six patients (with six involved hips) had an increase (mean, one centimeter; range, 0.5 to two centimeters) in the limb-length discrepancy after the index revision. In fourteen hips, a hip center that had been high preoperatively was moved distally, by the use of a larger acetabular component. These fourteen hips were associated with a mean of 1.6 centimeters of shortening preoperatively and 0.3 centimeter of greater length postoperatively.
In twenty hips, the hip center was moved proximally, a mean of 7.4 millimeters. Twelve of these hips had had a high hip center preoperatively, and eight had not. Of these twenty hips, thirteen had had a femoral revision. Ten of these thirteen hips were associated with an increase in limb length.
This study involved a somewhat unique group of patients in whom reconstruction was particularly challenging. Patients such as ours are commonly included in general studies of acetabular revision in which the results of both difficult and more standard procedures are assessed together. We elected to study these patients as a separate group in order to outline the challenges that are involved in their treatment with reconstruction as well as to highlight the successful long-term outcome (mean duration of follow-up, 10.4 years) that can be achieved without the use of a structural allograft or a protrusio cage in patients who have major loss of acetabular bone.
Many of our patients had had multiple operations on the hip before the index revision, often because of congenital dysplasia or dislocation of the hip. Thirty-nine (85 percent) of the hips had stage-IIIA or worse acetabular bone stock, thirty (65 percent) had acetabular osteolysis, and thirty-four (74 percent) already had a high hip center preoperatively. Forty-two hips (91 percent) needed extensive grafting of defects, such that the mean contact with host bone was only 79 percent. Additional pelvic bone was removed from fifteen hips to relieve impingement.
Despite the difficult circumstances that these deficient acetabula presented, excellent mechanical stability was achieved without the use of a structural graft, a protrusio cage, or cement. Rather than attempting to augment the bone stock with a bulk graft, we elected to form a hemisphere from the existing acetabular cavity and to carefully fill voids with particulate graft. The rate of mechanical loosening was 6 percent (two of thirty-six hips) at a mean of 10.4 years. These results appear to be substantially better than the short and intermediate-term results of acetabular revision with cement in these types of hips; the reported rates of repeat revision after such procedures have ranged from 5 to 51 percent4,10,23,31 in series ranging in size from sixty to 166 hips and after durations of follow-up of two to fourteen years.
Two hips in our study had a continuous radiolucent line. However, unlike the situation with cemented acetabular components17, the meaning of a continuous line at the implant-bone interface of a component that was inserted without cement is unclear. In a study of retrieved implants, Sumner et al. found that radiolucent areas had a higher proportion of fibrous ingrowth, as opposed to osseous ingrowth, when compared with interface zones without radiolucency41. Because localized areas of osseous ingrowth may not be visible on radiographs, and an acetabular component may be stable despite extensive radiolucency, we consider migration to be the most reliable indicator of loosening of an acetabular component that was inserted without cement.
In both of the hips in which the acetabular component loosened, it failed by migrating through the medial wall of the acetabulum. Both were small-diameter shells without contact with the pubis inferiorly. Support against the pubis can often be achieved with a so-called jumbo acetabular component in the presence of a large acetabular cavity, and when such a reconstruction is possible it is our preferred technique. In some hips in this study, however, the anteroposterior diameter of the acetabulum could not accommodate a large component. With the two exceptions already mentioned, high placement of the acetabular component in the superior segment of the acetabular recess did not prevent the formation of a stable bone-implant interface.
Acetabular reconstruction with a high hip center requires a number of other considerations. The most common criticism of the high hip center has been that it results in suboptimum biomechanics of the hip. Johnston et al. made a mathematical model of the hip joint and reported that a two-centimeter displacement of the hip center superiorly, posteriorly, and laterally results in a 22 percent increase in the resultant joint force18. Brand and Pedersen, however, showed that isolated superior displacement of the hip center increases the hip-joint force by only 5 percent3. In an experimental model of hip-joint forces, with testing of both single-limb stance and stair-climbing, Doehring et al. confirmed that movement of the hip center only superiorly does not affect the magnitude or direction of joint force10. In a computer simulation of pelvic anatomy and the normal relationships between muscle length and direction, Delp and Maloney demonstrated that, if the hip center is moved two centimeters superiorly, the force-generating capability of the hip abductors is reduced by 44 percent9. The authors, however, did not correct for the retensioning of the abductor muscles that occurs when the limb-shortening is corrected by the design of the femoral component or by advancement of the greater trochanter. In a second simulation, an allowance was made for restoration of the abductor tension, and the authors estimated that movement of the hip center two centimeters superiorly reduced the abductor strength by 18 percent43.
In the present study, the hip center began in a proximal location in thirty-four (74 percent) of the forty-six hips; thus, the high hip center was not created acutely. In twenty hips, the hip center was slightly more proximal (a mean of 7.4 millimeters), but it was also moved medially rather than laterally. The hip center was actually moved distally in nineteen hips, and it was moved laterally in this group. Accordingly, fewer patients had a positive Trendelenburg sign after the acetabular revision than before it, although a relatively high proportion (81 percent) had some limp. This finding is not surprising, given the number of previous operations that had been performed, the difficulty of the index acetabular revision, and the large number of patients who had underlying congenital hip disease.
Two other issues inherent to the high hip center are impingement and trochanteric nonunion. Careful intraoperative assessment of impingement generally allows the surgeon to correct this problem. However, even so, the rate of dislocation in this series was 11 percent (five of forty-six), which is higher than the rates of 6 percent (eight of 129) and 7 percent (four of sixty) reported in other series of revisions of similar components20,38. The rate of trochanteric nonunion was 11 percent (five of forty-four), and a previous osteotomy or a nonunion after a previous osteotomy put the patient at special risk for nonunion of the greater trochanter. This finding is different from that reported by Bal et al.1. Nonunion or migration of the trochanter did not necessitate a reoperation in our series.
Another important problem that has been identified in many studies of revision is infection. The overall rate of infection in our series was 7 percent (three of forty-six). Silverton et al. reported a similar rate of 5 percent (six of 129) in a series of acetabular revisions, with only one of the infections developing in the immediate postoperative period38. Berry and Müller reported a rate of infection of 12 percent (five of forty-two) in their series of acetabular reconstructions with use of a protrusio cage2. Two of the infections in the present series followed a reoperation on the femoral side. The negative effect of infection on these patients was very high. All three patients had removal of the implant, and two of them needed a hip disarticulation. The third patient did not have the components reimplanted because of insufficient bone stock.
The type of reconstruction described in this report has a number of advantages. Structural bone-grafting is avoided, and an acetabular component fixed without cement is supported by host bone, with particulate graft used to fill defects. The largest possible acetabular component should be used, in order to place the hip center as far distally as possible and to maximize the surface area for potential bone ingrowth as well as the thickness of the polyethylene. In many patients, however, the dimensions of the pelvis dictate the use of a smaller-diameter component, which also provides a reliable reconstruction. If the hip center is moved proximally, limb length should be restored with the femoral reconstruction and proper tension of the abductor muscles should be restored with trochanteric advancement.
Our experience with reconstruction involving a high hip center and use of a hemispherical acetabular component fixed with screws and without cement is clearly superior to our ten-year experience with the cementing of polyethylene components into bulk allografts. However, we have not used bilobed acetabular components or elliptical components, and we have used few protrusio cages, so we are unable to compare our experience with a high hip center with those types of procedures.
Bal, B. S.; Maurer, B. T.; and Harris, W. H.: Trochanteric union following revision total hip arthroplasty. J. Arthroplasty,13: 29-33, 1998.1329
1998
[PubMed]
Berry, D. J., and Müller, M. E.: Revision arthroplasty using an anti-protrusio cage for massive acetabular bone deficiency. J. Bone and Joint Surg.,74-B(5): 711-715, 1992.74-B(5)711
1992
Brand, R. A., and Pedersen, D.: Hip forces resulting from altered hip joint location revisited. Read at the Annual Meeting of the Hip Society, Scottsdale, Arizona, Sept. 20, 1990.
Callaghan, J. J.; Salvati, E. A.; Pellicci, P. M.; Wilson, P. D., Jr.; and Ranawat, C. S.: Results of revision for mechanical failure after cemented total hip replacement, 1979 to 1982. A two to five-year follow-up. J. Bone and Joint Surg.,67-A: 1074-1085, Sept. 1985.67-A1074
1985
Cameron, H. U., and Jung, Y. B.: Revision of the problem acetabulum with a "big head" bipolar prosthesis. Acta Orthop. Belgica,53: 467-471, 1987.53467
1987
Dearborn, J. T., and Harris, W. H.: Acetabular revision arthroplasty using "jumbo" cementless components. An average seven-year follow-up study. Personal communication, 1997.
Dearborn, J. T., and Murray, W. R.: Arthopor 2 acetabular component with screw fixation in primary hip arthroplasty: a 7- to 9-year follow-up study. J. Arthroplasty,13: 299-310, 1998.13299
1998
[PubMed]
DeLee, J. G., and Charnley, J.: Radiological demarcation of cemented sockets in total hip replacement. Clin. Orthop.,121: 20-32, 1976.12120
1976
[PubMed]
Delp, S. L., and Maloney, W.: Effects of hip center location on the moment-generating capacity of the muscles. J. Biomech.,26: 485-499, 1993.26485
1993
[PubMed]
Doehring, T. C.; Rubash, H. E.; Shelley, F. J.; Schwendeman, L. J.; Donaldson, T. K.; and Navalgund, Y. A.: Effect of superior and superolateral relocations of the hip center on hip joint forces. An experimental and analytical analysis. J. Arthroplasty,11: 693-703, 1996.11693
1996
[PubMed]
Dorr, L. D., and Wan, Z.: Ten years of experience with porous acetabular components for revision surgery. Clin. Orthop.,319: 191-200, 1995.319191
1995
[PubMed]
Garbuz, D.; Morsi, E.; and Gross, A. E.: Revision of the acetabular component of a total hip arthroplasty with a massive structural allograft. Study with a minimum five-year follow-up. J. Bone and Joint Surg.,78-A: 693-697, May 1996.78-A693
1996
Goodman, S. B.; Adler, S. J.; Fyhrie, D. P.; and Schurman, D. J.: The acetabular teardrop and its relevance to acetabular migration. Clin. Orthop.,236: 199-204, 1988.236199
1988
[PubMed]
Gruen, T. A.; McNeice, G. M.; and Amstutz, H. C.: "Modes of failure" of cemented stem-type femoral components. A radiographic analysis of loosening. Clin. Orthop.,141: 17-27, 1979.14117
1979
[PubMed]
Haentjens, P.; de Boeck, H.; Handelberg, F.; Casteleyn, P.-P.; and Opdecam, P.: Cemented acetabular reconstruction with the Müller support ring. A minimum five-year clinical and roentgenographic follow-up study. Clin. Orthop.,290: 225-235, 1993.290225
1993
[PubMed]
Harris, W. H.: Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J. Bone and Joint Surg.,51-A: 737-755, June 1969.51-A737
1969
Hodgkinson, J. P.; Shelley, P.; and Wroblewski, B. M.: The correlation between the roentgenographic appearance and operative findings at the bone-cement junction of the socket in Charnley low friction arthroplasties. Clin. Orthop.,228: 105-109, 1988.228105
1988
[PubMed]
Johnston, R. C.; Brand, R. A.; and Crowninshield, R. D.: Reconstruction of the hip. A mathematical approach to determine optimum geometric relationships. J. Bone and Joint Surg.,61-A: 639-652, July 1979.61-A639
1979
Kavanagh, B. F.; Ilstrup, D. M.; and Fitzgerald, R. H., Jr.: Revision total hip arthroplasty. J. Bone and Joint Surg.,67-A: 517-526, April 1985.67-A517
1985
Lachiewicz, P. F., and Hussamy, O. D.: Revision of the acetabulum without cement with use of the Harris-Galante porous-coated implant. Two to eight-year results. J. Bone and Joint Surg.,76-A: 1834-1839, Dec. 1994.76-A1834
1994
Livermore, J.; Ilstrup, D.; and Morrey, B.: Effect of femoral head size on wear of the polyethylene acetabular component. J. Bone and Joint Surg.,72-A: 518-528, April 1990.72-A518
1990
McGrory, B. J.; Bal, B. S.; and Harris, W. H.: Trochanteric osteotomy for total hip arthroplasty: six variations and indications for their use. J. Am. Acad. Orthop. Surgeons,4: 258-267, 1996.4258
1996
Marti, R. K.; Schüller, H. M.; Besselaar, P. P.; and Vanfrank Haasnoot, E. L.: Results of revision of hip arthroplasty with cement. A five to fourteen-year follow-up study. J. Bone and Joint Surg.,72-A: 346-354, March 1990.72-A346
1990
Mulroy, R. D., Jr., and Harris, W. H.: Failure of acetabular autogenous grafts in total hip arthroplasty. Increasing incidence: a follow-up note. J. Bone and Joint Surg.,72-A: 1536-1540, Dec. 1990.72-A1536
1990
Murray, W. R.: Salvage of acetabular insufficiency with bipolar prostheses. In The Hip. Proceedings of the Twelfth Open Scientific Meeting of the Hip Society, pp. 296-311. St. Louis, C. V. Mosby, 1984.
Namba, R., and Scott, R. D.: Acetabular revision with the bipolar prosthesis and particulate bone grafting. Sem. Arthroplasty,4: 87-91, 1993.487
1993
Namba, R. S.; Janku, G. V.; and Murray, W. R.: Reconstruction of major segmental acetabular defects with a porous coated oblong component. Orthop. Trans.,20: 899, 1996-1997.20899
1996-1997
Oakeshott, R. D.; Morgan, D. A. F.; Zukor, D. J.; Rudan, J. F.; Brooks, P. J.; and Gross, A. E.: Revision total hip arthroplasty with osseous allograft reconstruction. A clinical and roentgenographic analysis. Clin. Orthop.,225: 37-61, 1987.22537
1987
[PubMed]
Papagelopoulos, P. J.; Lewallen, D. G.; Cabanela, M. E.; McFarland, E. G.; and Wallrichs, S. L.: Acetabular reconstruction using bipolar endoprosthesis and bone grafting in patients with severe bone deficiency. Clin. Orthop.,314: 170-184, 1995.314170
1995
[PubMed]
Possai, K. W.; Dorr, L. D.; and McPherson, E. J.: Metal ring supports for deficient acetabular bone in total hip replacement. In Instructional Course Lectures, American Academy of Orthopaedic Surgeons. Vol. 45, pp. 161-169. Rosemont, Illinois, American Academy of Orthopaedic Surgeons, 1996.
Raut, V. V.; Siney, P. D.; and Wroblewski, B. M.: Revision of the acetabular component of a total hip arthroplasty with cement in young patients without rheumatoid arthritis. J. Bone and Joint Surg.,78-A: 1853-1856, Dec. 1996.78-A1853
1996
Rosson, J., and Schatzker, J.: The use of reinforcement rings to reconstruct deficient acetabula. J. Bone and Joint Surg.,74-B(5): 716-720, 1992.74-B(5)716
1992
Russotti, G. M., and Harris, W. H.: Proximal placement of the acetabular component in total hip arthroplasty. A long-term follow-up study. J. Bone and Joint Surg.,73-A: 587-592, April 1991.73-A587
1991
Schmalzried, T. P., and Harris, W. H.: The Harris-Galante porous-coated acetabular component with screw fixation. Radiographic analysis of eighty-three primary hip replacements at a minimum of five years. J. Bone and Joint Surg.,74-A: 1130-1139, Sept. 1992.74-A1130
1992
Schutzer, S. F., and Harris, W. H.: Trochanteric osteotomy for revision total hip arthroplasty. 97% union rate using a comprehensive approach. Clin. Orthop.,227: 172-183, 1988.227172
1988
[PubMed]
Schutzer, S. F., and Harris, W. H.: High placement of porous-coated acetabular components in complex total hip arthroplasty. J. Arthroplasty,9: 359-367, 1994.9359
1994
[PubMed]
Shinar, A. A., and Harris, W. H.: Bulk structural autogenous grafts and allografts for reconstruction of the acetabulum in total hip arthroplasty. Sixteen-year-average follow-up. J. Bone and Joint Surg.,79-A: 159-168, Feb. 1997.79-A159
1997
Silverton, C. D.; Rosenberg, A. G.; Sheinkop, M. B.; Kull, L. R.; and Galante, J. O.: Revision of the acetabular component without cement after total hip arthroplasty. A follow-up note regarding results at seven to eleven years. J. Bone and Joint Surg.,78-A: 1366-1370, Sept. 1996.78-A1366
1996
Stock, J. R.; Athanasoulis, C. A.; Harris, W. H.; Waltman, A. C.; Novelline, R. A.; and Greenfield, A. J.: Transcatheter embolization for the control of wound hemorrhage following hip surgery. J. Bone and Joint Surg.,62-A: 1000-1003, Sept. 1980.62-A1000
1980
Stockl, B.; Beerkotte, J.; Krismer, M.; Fischer, M.; and Bauer, R.: Results of the Muller acetabular reinforcement ring in revision arthroplasty. Arch. Orthop. and Trauma Surg.,116: 55-59, 1997.11655
1997
Sumner, D. R.; Jasty, M.; Jacobs, J. J.; Urban, R. M.; Bragdon, C. R.; Harris, W. H.; and Galante, J. O.: Histology of porous-coated acetabular components. 25 cementless cups retrieved after arthroplasty. Acta Orthop. Scandinavica,64: 619-626, 1993.64619
1993
Tanzer, M.; Drucker, D.; Jasty, M.; McDonald, M.; and Harris, W. H.: Revision of the acetabular component with an uncemented Harris-Galante porous-coated prosthesis. J. Bone and Joint Surg.,74-A: 987-994, Aug. 1992.74-A987
1992
Vasavada, A. N.; Delp, S. L.; Maloney, W. J.; Schurman, D. J.; and Zajac, F. E.: Compensating for changes in muscle length in total hip arthroplasty. Effects on the moment generating capacity of the muscles. Clin. Orthop.,302: 121-133, 1994.302121
1994
[PubMed]
Zehntner, M. K., and Ganz, R.: Midterm results (5.5—10 years) of acetabular allograft reconstruction with the acetabular reinforcement ring during total hip revision. J. Arthroplasty,9: 469-479, 1994.9469
1994
[PubMed]