Harris-Galante-I porous-coated acetabular components are hemispherical prostheses of titanium alloy with pure titanium fiber-metal mesh for bone ingrowth6,20. They have multiple holes to allow supplemental fixation with either 4.5 or 5.1-millimeter titanium-alloy screws. The modular polyethylene liner is secured with three pairs of metal tines located at the rim of the prosthesis.
One hundred and fifty-four total hip arthroplasties were performed by the senior one of us (P. F. L.) with the Harris-Galante-I porous-coated acetabular component in 124 patients between December 1984 and December 1989. Ten patients, who had eleven well functioning hip prostheses, died of unrelated causes before the five-year evaluation; five hips in four patients were lost to follow-up; and two replacements in two patients were removed because of late infection of the hip at four and five years. One hundred and thirty-six hips in 108 patients had complete radiographic and clinical follow-up and formed the basis of this study.
The mean duration of follow-up was seven years, with a range of five to ten years. There were seventy-two women (ninety-three hips) and thirty-six men (forty-three hips). The mean weight of the patients was seventy-five kilograms (range, fifty-two to 100 kilograms). Fifty (46 per cent) of the patients were considered obese, as defined by a weight that was greater than 119 per cent of the ideal body weight according to the 1983 Metropolitan Life Insurance Company tables25. The preoperative diagnosis was osteoarthrosis in fifty-seven hips (42 per cent), rheumatoid arthritis in thirty-nine (29 per cent), osteonecrosis in twenty-nine (21 per cent), post-traumatic osteoarthrosis in six (4 per cent), and various other diagnoses in five (4 per cent).
A posterior approach was used in 132 hips, and a transtrochanteric approach was used in four. The acetabulum was reamed progressively with hemispherical reamers to the medial wall to obtain bleeding cancellous or subchondral bone. The outer diameter of the acetabular component that was implanted was the same as the diameter of the final reamer that was used. Unlike the technique described by Schmalzried and Harris20, the last reamer was used only briefly at the acetabular rim. A stable press-fit was obtained when the prosthesis was implanted with the cup-holder. Additional fixation was routinely obtained with three to six screws (mean, four screws), usually directed superiorly and posteriorly. A screw was routinely inserted in the polar hole of the prosthesis and directed toward the iliopectineal line of the pelvis.
Autogenous grafts from the femoral head were secured with screws in eighteen hips with segmental acetabular defects. The preoperative diagnosis was protrusio acetabuli secondary to rheumatoid arthritis in seven of these hips, osteoarthrosis in six, congenital dysplasia in two, post-traumatic osteoarthrosis in two, and osteonecrosis in one. Morselized cancellous autogenous graft was used to fill empty holes in the component and in arthritic cysts. A polyethylene liner with an elevated rim was not used in these hips.
One hundred hips had a Harris-Galante femoral component inserted without cement, twenty-eight had a Triad femoral component (Johnson and Johnson, Braintree, Massachusetts) inserted with cement, and eight had a Precoat femoral component (Zimmer) inserted with cement. The decision to use cement to fix the femoral component was based on the age of the patient and the expected level of activity. The mean age of the patients (100 hips) in whom the femoral component was inserted without cement was forty-five years (range, eighteen to sixty-eight years), and the mean age of the patients (thirty-six hips) in whom it was inserted with cement was sixty-five years (range, forty-four to seventy-eight years). A twenty-eight-millimeter chromium-cobalt femoral head was used in 130 hips. This femoral head size had previously been the only one used by the senior one of us so it was arbitrarily selected for most of the hips. A thirty-two-millimeter head was used in five hips in which a very large acetabular component was implanted, and a twenty-two-millimeter head was used in one hip in order to provide more than six millimeters of polyethylene thickness in a patient who had a very small acetabulum.
The hips were prospectively evaluated clinically with the Harris hip score. Standardized anteroposterior radiographs of the pelvis, centered over the pubis, were made for all patients postoperatively at six weeks, at three months, at six months, and yearly thereafter. The angle of abduction of the acetabular component was measured on the anteroposterior radiograph as the angle formed by a line drawn tangential to the face of the acetabulum and a line drawn tangential to the inferior margins of the ischial tuberosities. The postoperative true lateral radiograph was used to assess relative anteversion of the component27. Radiographic evaluation of the acetabular component on the anteroposterior radiograph of the pelvis included measurement of radiolucent and radiodense lines in the three zones described by DeLee and Charnley. Radiolucent lines were not analyzed on the lateral radiographs. An acetabular component was considered loose if there was at least two millimeters of migration or a change of at least 4 degrees in the angle of abduction. Vertical and horizontal migration were evaluated with the method of Massin et al. The thickness of the polyethylene was assessed, with one micrometer being the shortest distance from the outer edge of the femoral head to the outer wall of the acetabulum. The measured size of the femoral head was used to correct for magnification. Wear of the polyethylene was measured for the 131 hips that did not have the liner replaced during revision of a loose femoral component (Fig. 1). Clinical evaluation with use of the Harris hip score was performed for 129 hips in which the femoral component was not revised.
Clinical Analysis
The Harris hip score was used to evaluate ninety-four hips (eighty patients) with a Harris-Galante femoral component that had been inserted without cement. Sixty-nine hips (fifty-six patients) were rated as excellent, eighteen (seventeen patients) were rated as good, four (four patients) were rated as fair, and three (three patients) were rated as poor at the most recent follow-up evaluation. Fifty-six patients (seventy hips) had no pain, twenty-two (twenty-two hips) had slight pain, and two (two hips) had mild pain that necessitated the occasional use of non-narcotic analgesics. Seventy-six patients (eighty-nine hips) had no or a slight limp, three (four hips) had a moderate limp, and one (one hip) was unable to walk. Sixty-eight patients (eighty-two hips) did not use any supports, six (six hips) used a cane on long walks, two (two hips) used a crutch, three (three hips) used two canes, and one (one hip) was unable to walk because of severe rheumatoid arthritis in other joints. Forty-eight patients (fifty-five hips) were able to walk an unlimited distance, twelve (fourteen hips) could walk as far as six blocks, fourteen (nineteen hips) could walk two to three blocks, five (five hips) walked indoors only, and one (one hip) was unable to walk as a result of severe rheumatoid arthritis in other joints.
Of the thirty-five hips (twenty-eight patients) in which the femoral component had been inserted with cement and that were evaluated with the Harris hip score, fourteen (ten patients) were rated as excellent, twelve (ten patients) were rated as good, four (three patients) were rated as fair, and five (five patients) were rated as poor at the most recent follow-up evaluation. Nineteen patients (twenty-six hips) had no pain, seven (seven hips) had slight pain, and two (two hips) had mild pain that necessitated the occasional use of non-narcotic analgesics. Fifteen patients (twenty-two hips) had no limp, five (five hips) had a slight limp, three (three hips) had a moderate limp, two (two hips) had a severe limp, and three (three hips) were unable to walk. Eleven patients (fifteen hips) did not use any supports, six (seven hips) used a cane for long walks, four (four hips) used a cane full-time, two (two hips) used a crutch, two (four hips) used two canes, and three (three hips) were unable to walk because of severe medical problems. Ten patients (thirteen hips) were able to walk an unlimited distance, three (four hips) were able to walk as far as six blocks, ten (thirteen hips) were able to walk two to three blocks, two (two hips) walked indoors only, and three (three hips) were unable to walk because of severe medical problems.
Reoperations and Complications
There were no intraoperative complications related to the placement of the screws. No acetabular component was revised for mechanical loosening, and no revisions were pending at the time of this review.
In two patients (who were excluded from the study, as already mentioned) who had a so-called hybrid total hip arthroplasty (an acetabular component inserted without cement and a femoral component inserted with cement), the well fixed Harris-Galante porous-coated acetabular component was removed at the time of a resection arthroplasty for late infection. One of the resections was performed at four years by the senior one of us, and the other was performed at five years at another institution. The histological evaluation3 of the acetabular component that was removed at our institution showed a relatively high total amount of bone ingrowth (8.4 per cent of total pore volume; apposition to bone, 23.8 per cent surface contact), particularly at areas near the screws and into the empty screw-holes.
One patient had a reoperation three and a half years after the initial operation because of recurrent posterior dislocations despite a satisfactory position of both components. A liner with a rim that was elevated 10 degrees was placed, and the posterior aspect of the capsule was repaired. There were no additional dislocations during the ensuing five years of follow-up.
Severe osteolysis of the ischium and the posterior acetabular wall developed at six and seven and a half years after the operation in two patients (two hips) managed with a porous-coated femoral component who had both sickle-cell disease and osteonecrosis. For both patients, oil-red-O stain21 was used to confirm the presence of particles of polyethylene debris in the tissue in the osteolytic lesions. Both patients were asymptomatic, but concerns regarding loss of bone and possible catastrophic failure led the senior one of us to explore the two hips. At the exploration, the acetabular components were found to be rigidly fixed and were left in place. The areas of osteolysis were filled with crushed cancellous allograft bone. The twenty-eight-millimeter femoral head was exchanged for a twenty-two-millimeter head in one patient and for a twenty-six-millimeter head in the other patient. A new, corresponding polyethylene liner was inserted in each patient. At one and two years after the reoperation, both patients had an excellent clinical result, with almost complete healing of the grafted areas.
During the three years before the most recent evaluation, five femoral components that had been inserted without cement and one cemented femoral component were revised because of loosening. The polyethylene liner was replaced in five of these hips at the time of the revision.
Radiographic Analysis
The mean angle of abduction of the acetabular component was 42 degrees (range, 30 to 55 degrees). All of the components had been placed in anteversion or neutral. At the most recent follow-up examination, there was no migration of any acetabular component. There were no bent or broken screws, and there was no osteolysis associated with the screws or screw-holes. In one hip, asymptomatic breakage of one of the peripheral locking tines was noted four years postoperatively. There was no additional breakage of tines or dissociation of the polyethylene liner during the ensuing four years. A radiodense line was seen in zone 1 in nineteen hips, in zone 2 in five hips, and in zone 3 in sixteen hips. A non-progressive radiolucent line was seen in zone 1 only in twenty-one hips, in zone 2 only in three hips, and in zone 3 only in ten hips. Such a line was seen in two adjacent zones in two hips and in two non-adjacent zones in four hips. No acetabular component was surrounded by a radiolucent line in all three zones.
Of the 131 hips that had not had the polyethylene liner replaced during revision of a loose femoral component, 100 had less than one millimeter of linear wear of the polyethylene, twenty-eight had one to two millimeters, and three had more than two millimeters. The mean rate of linear wear in all 131 hips was 0.1 millimeter per year.
As mentioned, osteolysis of the ischium and posterior acetabular wall was seen in two of three hips that had more than two millimeters of wear of the polyethylene (Figs. 2-A, 2-B, and 2-C). Both had measurable eccentricity of the femoral head within the acetabular component. In one of the hips, 4.2 millimeters of wear was measured radiographically at seven and a half years. Measurement of the retrieved liner from the other hip showed 3.7 millimeters of linear wear of the polyethylene.
Since December 1984, we have prospectively evaluated the clinical and radiographic results associated with 154 Harris-Galante-I porous-coated acetabular components fixed with screws in 124 patients. Only five hips were lost to follow-up during this study, leaving 149 (97 per cent) with a known clinical result. To our knowledge, this is the largest study, with the longest follow-up (mean, seven years; range, five to ten years), of this acetabular component. Clinically, 113 (88 per cent) of the 129 hips that did not have revision of the femoral component had a good or excellent result. The eight poor clinical results were not related to loosening of the component but to osteoarthrosis or rheumatoid arthritis in other joints or to severe medical problems that limited walking. This study is important because of recent controversy regarding the method of implantation and the use of screws for fixation of acetabular components inserted without cement.
The successful results reported for this porous-coated acetabular component are probably related in part to the method of implantation. The superiority of screws compared with pegs or spikes for achieving initial stability of acetabular components inserted without cement has been shown in a cadaver pelvis model11. In addition, compression, obtained only through the use of screws, may improve contact at the bone-prosthesis interface and increase the amount of bone ingrowth. Concerns regarding metal-fretting and the creation of access channels for polyethylene debris have led other surgeons to recommend the avoidance of screws and to obtain fixation only by a so-called press-fit into an underreamed acetabulum. However, both Schwartz et al. and MacKenzie et al. showed a high prevalence of acetabular fractures and decreased surface contact, especially in the polar region, with this technique. Our operative technique of using the reamer briefly at the rim allowed us to gain a stable press-fit without the creation of gaps at the polar dome. The addition of multiple screws improved stability and, it is hoped, bone ingrowth. One of the acetabular components retrieved during a resection arthroplasty for infection at four years showed extensive ingrowth of bone, predominantly into the porous coating adjacent to the screws as well as into empty screw-holes3. There were no operative or postoperative complications associated with the screws at five to ten years after the operation.
Pelvic osteolysis was reported by Santavirta et al. and by Maloney et al.15 after total hip arthroplasty performed without cement. Two young, active patients in the present study, both of whom had sickle-cell disease and osteonecrosis, had severe wear of the polyethylene that was first detected on the radiographs made at six years. Osteolytic lesions developed in the ischium and the posterior acetabular rim at six and seven and a half years. The acetabular components were found to be rigidly fixed and were left in place at the time of exploration; the lesions were packed with allograft bone. The prosthetic femoral heads were exchanged for smaller components so that a thicker polyethylene liner could be placed. In order to use at least an eight-millimeter thickness of polyethylene for the acetabular cup, we now insert a femoral head component with a diameter of twenty-two millimeters when an acetabular component with an outer diameter of less than fifty-four millimeters is used.
The technique of measuring polyethylene wear used in this study was developed by Maloney14 and is a modification of the technique developed by Griffith et al. However, since only one retrieved polyethylene liner was measured directly, we are unable to comment on the absolute accuracy of this technique. There were too few femoral components with a head size of other than twenty-eight millimeters to confirm the finding of Livermore et al. that thirty-two-millimeter heads have more polyethylene wear than do twenty-eight or twenty-two-millimeter heads. The over-all rate of linear wear of 0.1 millimeter per year in the present study is comparable with that previously reported28.
The results of arthroplasty with the Harris-Galante-I porous-coated acetabular component with screw fixation are a notable improvement compared with those reported for several other types of acetabular components inserted without cement2,5,9. However, the results with acetabular components inserted with cement are the standard against which new methods of fixation should be judged. Stauffer reported loosening, at ten years, of 11 per cent (twenty-six) of 231 Charnley components that had been inserted with so-called original cementing techniques. Schulte et al. reported that 23 per cent (twenty-two) of ninety-four cemented Charnley components were definitely loose or had been revised at twenty years. The results with cemented acetabular components in younger patients are not comparable. Ballard et al. reported that 36 per cent (fifteen) of forty-two cemented acetabular components were definitely loose or had been revised at ten years in patients who were less than fifty years old.
The results with the Harris-Galante-I porous-coated acetabular component were excellent despite the use of the Harris-Galante femoral component, which is associated with a high prevalence of subsidence10, and the Triad femoral component, which is associated with a high prevalence of debonding and loosening26. While the results at a mean of seven years are encouraging, other acetabular components inserted without cement or with screw fixation have been associated with very high rates of adjacent radiolucent lines, migration, osteolysis, and revision after similar durations of follow-up2,5,9. As cemented acetabular components are generally reported to have reasonable fixation at five to ten years and then increasing rates of failure, at least ten years of follow-up of Harris-Galante-I components is required to determine if this method of fixation is definitely better than fixation with cement.