 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. The Hemispherical Harris-Galante Acetabular Cup, Inserted without Cement. The Results of an Eight to Eleven-Year Follow-up of One Hundred and Sixty-Eight Hips* , HILLERØD, DENMARKInvestigation performed at the Department of Orthopaedic Surgery, Hillerød Hospital, Hillerød
We studied the results for 168 available hips from a series of 324 consecutive primary total hip arthroplasties that had been performed with insertion of a Harris-Galante-I acetabular component without cement. The acetabulum had been reamed in a so-called line-to-line manner, and the cup had been fixed with one to four screws. A femoral component with a modular alumina-ceramic head had been inserted with cement in all hips. The median duration of follow-up was 112 months (range, 101 to 131 months). Of the original 324 hips, 109 could not be included in the clinical and radiographic follow-up because the patients had died and thirty could not be included because the patients were not available for examination. Seventeen hips had had a revision of the acetabular cup: five, because of infection; five, because of dislocation; three, because of aseptic loosening; and four, because of technical failure. This left 168 hips for clinical and radiographic follow-up; of these, fifteen had had a revision of the femoral component only. Of the remaining 153 hips, which had not had a revision, 147 (96 percent) were considered by the patient to have a satisfactory, good, or excellent result. One hip was found to have a loose cup on radiographic evaluation and was therefore considered to have failed, but the clinical function was good. We concluded that, with an overall rate of aseptic loosening of 1 percent (four of 324) after an intermediate (ten-year) duration of follow-up, use of this cup has good results.
Fixation of the acetabular cup with cement has been a standard technique for total hip arthroplasty for many years. Excellent results have been reported in short-term follow-up studies; however, there has been an increased frequency of aseptic loosening in long-term follow-up studies. Improved cementing techniques have increased the long-term durability of fixation of the femoral component, whereas loosening of the acetabular component continues to be a major long-term problem6,13,16,24,28. In an attempt to improve the results, acetabular components have increasingly been inserted without cement15. The Harris-Galante-I acetabular cup (Zimmer, Warsaw, Indiana) was first used in our department in July 1985. Since then, all primary total hip arthroplasties, as well as revision operations, have been performed with use of this cup. The results of primary7,9,11,12,22,23,27,29 and revision19,26 hip arthroplasties with insertion of a Harris-Galante-I cup in combination with a modular metallic head have been studied. Favorable results with use of this cup also were documented in the Norwegian Arthroplasty Register study4. The present study is based on the results of clinical and radiographic follow-up after our first 324 consecutive primary total hip arthroplasties with use of the Harris-Galante-I cup, performed between July 1985 and December 1987. All cups were matched with a femoral component with a modular alumina-ceramic head, inserted with cement.
The Harris-Galante-I acetabular component is a hemispherical titanium cup. It is fully covered by clinched and entangled commercially pure titanium fibers in a fiber-metal pad. The pad has a thickness of 1.6 millimeters, a density of 50 percent, and an average pore size of 350 micrometers. All cups were inserted with so-called line-to-line reaming, meaning that the diameter of the cup matched the size of the largest reamer that was used. Supplementary fixation was achieved with one to four self-tapping cancellous-bone screws. The surgeon decided on the number of screws at the time of the operation. Three hundred and forty-seven primary total hip replacements were performed in 321 patients (175 women and 146 men). The median age of the patients at the time of the index operation was sixty-seven years (range, nineteen to eighty-six years). Thirty-seven patients were less than fifty years old. In 324 hips, the Harris-Galante-I cup was inserted without cement and was matched with a femoral component that was inserted with cement; a Protazul femoral component (Zimmer) was used in 313 of these hips, and an LMT femoral component (Biomet, Warsaw, Indiana) was used in eleven. An LMT femoral component was inserted without cement in the remaining twenty-three hips, which were excluded from additional analysis in order to provide a more uniform population. (These twenty-three hips had good survival of both the cup and the stem.) In all hips, a modular alumina-ceramic head (Feldmühle, Plochingen, Germany) corresponding to the taper of the femoral component was used. Autologous bone graft was used to reconstruct eight acetabula. Only morseled bone was used in two, and a structural femoral head graft was used in six. The ratio of coverage by the graft to coverage by the host bone was not recorded. At the time of the latest follow-up, none of these eight hips had had a revision or were scheduled to have one. The operations were performed in an ultra-clean-air environment, with use of antibiotic prophylaxis and a posterolateral approach. The median duration of the operation was 101 minutes (range, fifty-five to 200 minutes). Full weight-bearing with use of crutches was allowed on the first or second postoperative day. The median duration of follow-up was 112 months (range, 101 to 131 months). At the time of the latest follow-up, a total of seventeen cups had been revised and 106 patients (113 hips) had died. (Four of the revised hips were in patients who died later.) Two patients (two hips) could not be located. Of the remaining 179 patients (196 hips), twenty-three (twenty-eight hips) refused to participate in the study. Thus, 156 patients (eighty-four women and seventy-two men) with a total of 168 hips returned for radiographic and clinical examination. The preoperative diagnosis was primary osteoarthritis in 136 hips (81 percent), osteonecrosis following fracture of the femoral neck in fourteen hips (8 percent), idiopathic osteonecrosis of the femoral head in seven hips (4 percent), and rheumatoid arthritis in four hips (2 percent); the remaining seven hips (4 percent) had other indications for the operation. Eight patients who were seen at the latest follow-up evaluation had been less than fifty years old at the time of the index operation. In fifteen hips, the femoral component had been revised, without revision of the cup and liner, after the index operation; these hips were therefore evaluated as a separate group. Immediate postoperative radiographs were available for only fifty-nine of the 168 hips because of archival procedures. At the time of the latest follow-up, standardized anteroposterior and axial radiographs of the pelvis and hip, with a focal spot-to-film distance of one meter, were made and evaluated for radiolucent lines according to the method of DeLee and Charnley2, for migration according to the system of Massin et al.14, and for wear of the polyethylene liner according to the criteria of Latimer and Lachiewicz12. Ectopic bone formation was classified according to the system of Brooker et al.1. Clinically, the patients were evaluated according to the scoring system of Harris3 and with use of a separate questionnaire concerning function and the patient's evaluation of the result. Chi-square analysis was used to test the significance of the results. Standard Kaplan-Meier survival curves8 were constructed to estimate the overall risk of failure of the cup. The study was performed in accordance with the Helsinki II Declaration and was approved by the local ethical committee.
Clinical Outcome Ninety-eight patients (106 [69 percent] of the 153 hips) were able to walk more than one kilometer, and forty-four patients (forty-seven hips; 31 percent) could walk less than 500 meters. However, twenty-nine (62 percent) of these forty-seven hips were thought by the patient to have a good or excellent result. Of the fifteen hips that had had a revision of the femoral component without a revision of the acetabular component, eight were not painful, five were mildly or slightly painful, and two were more painful than they had been before the operation. Nine of the fifteen patients who had had a revision of only the femoral component were able to walk more than one kilometer. Of the 153 hips that had not had a revision, 147 (96 percent) were considered by the patient to have a satisfactory result (sixteen hips), a good result (seventy-seven hips), or an excellent result (fifty-four hips). The remaining six hips had a poor result in the patient's opinion. The median Harris hip score3 for the 153 hips was 86 points (range, 27 to 100 points) at the time of the latest follow-up. The score was considered excellent for seventy-two hips, good for thirty-one, fair for twenty-five, and poor for twenty-five. The fifteen hips that had had a revision of the femoral component only had a median Harris hip score of 84 points (range, 20 to 100 points).
Radiographic Evaluation
Examination of the pelvic bone close to the cup (areas without radiolucency) revealed findings ranging from a marked continuous radiodense line approximately one millimeter in thickness along the metallic border of the cup to completely homogeneous bone that was in intimate contact with the surface of the metallic mesh around the cup. No osteolysis was noted in association with the screws or screw-holes, and there were no other clear radiographic signs of osteolysis or osteopenia. Radiolucent lines were seen around ninety-one femoral components. Twenty of these components were associated with a continuous radiolucent line, and seven of the twenty had clearly failed according to radiographic criteria. Four of these hips were immediately scheduled to have a revision. Revision was not scheduled for the remaining three because of medical risk factors or old age; instead, these patients were managed with careful follow-up. The median Harris hip score3 for these seven hips was 87 points (range, 37 to 100 points). Twenty-nine hips had marked signs of osteopenia in the proximal aspect of the femur. One hundred and forty-three hips had no ectopic ossification. Thirteen hips were class II, twelve were class III, and none was class IV, according to the system of Brooker et al.1. Wear was technically difficult to measure because of insufficient demarcation of the ceramic head on the radiographs. The wear seen on the original postoperative radiographs could be compared with that noted on the most recent follow-up radiographs for only a few hips. The cup that was possibly radiographically loose and had been implanted in an extremely vertical position had clearly measurable wear of two millimeters in the weight-bearing zone in the most superior area. Only limited wear was detected in the remaining hips, but the head was not clearly demarcated and the wear could not be quantified. Overall, eighteen Harris-Galante-I cups were considered to have failed at the time of the latest follow-up (Fig. 1). Fifteen stems that had been inserted with cement had been revised, without revision of the cup. Fourteen stems had been replaced because of aseptic loosening, and one had been revised after a traumatic fracture of the femur.
Causes of Failure of the Cup We detected no relationship between aseptic loosening and the underlying diagnosis for which the index operation had been performed. We also did not find that the number of screws that had been used for supplemental fixation of the cup during the index operation predicted an increased or reduced risk of loosening. Eight (2 percent) of the 324 hips had at least one episode of dislocation. No malposition of the cup was found in three hips, and revision therefore was not indicated. Two of these three hips had one episode of dislocation, and one had two episodes. The remaining five hips had one to four dislocations and were revised (three, within the first three months after the index operation). Only the cup was replaced in four of these hips, whereas both the stem and the cup were replaced in one hip. After the revision, two hips had additional dislocations. The five revisions because of deep infection were performed at eighteen to sixty-seven months after the index operation. The cup was found to be well fixed and was removed with difficulty from four of the five hips. The femoral stem was loose in all five hips. The three revisions because of aseptic loosening were done at ten, eighteen, and sixty-nine months after the index operation. The revision because of displacement of the cup was performed at ten days postoperatively. The three revisions because of displacement of the polyethylene liner out of the titanium shell were done at forty-three, fifty-three, and 102 months. At the time of the revision, the femoral head was within the socket, the flanges were distorted, and the insert was tilted out and rotated. One cup had a fracture of the rim. The first two hips have been described previously21.
To our knowledge, this is the first ten-year follow-up study of the results of use of the Harris-Galante-I hemispherical acetabular cup in combination with an alumina-ceramic femoral head. Similarly good results have been described after use of this cup in combination with a chromium-cobalt head in studies with a shorter duration of follow-up4,7,9,11,12,23,29. The fact that 102 patients (109 hips) who did not have a revision died before the latest follow-up evaluation is a problem with the current study. However, we examined all hospital records, including those of the medical departments, as well as data from the national central patient registration system, to ensure that no revisions had been performed and no other technical problems had arisen in these hips before the deaths of the patients. One hundred and sixty-eight hips could be evaluated, and seventeen hips had failure (revision) of the cup. The radiographic follow-up also revealed one loose cup. This study included a total of sixteen total hip arthroplasties that had been performed because of rheumatoid arthritis. It has been questioned whether components should be inserted without cement in such patients18. On the basis of the rate of revision and the radiographic findings, we do not believe that these cups are contraindicated in patients who have rheumatoid arthritis. However, the group was small, and additional studies are necessary. In the past, implants without cement were mostly used for younger patients, whereas cups were inserted with cement in older patients. In the present study, a hybrid prosthesis (a femoral component inserted with cement and a Harris-Galante-I cup inserted without cement) was used irrespective of the age of the patient. No association between an older age and loosening of the cup was detected. Only one patient had early displacement of the cup, which was probably due to a technical error. None of the patients had any signs of pelvic fracture or neurovascular complications. If the patients who died or were lost to follow-up are excluded, our findings show that, of the total population of 213 hips (195 patients) that were operated on, 147 (69 percent) had at least a satisfactory result in the opinion of the patient. Analgesic drugs were needed for thirty-six (24 percent) of the 153 hips that were examined. Fifty-three (32 percent) of the 168 hips that were evaluated were in patients who were able to walk less than 500 meters. However, this fact should be considered together with the fact that thirty-two (60 percent) of these less mobile hips were deemed by the patient to have a good or excellent result. It is our impression that the restricted walking distance was due to general physical weakness caused by old age or other diseases rather than to technical problems with the prosthesis. Thirty-nine (74 percent) of the fifty-three less mobile hips were in patients who had at least one concomitant disabling disease compared with seventy-seven (67 percent) of the more mobile hips. The Harris hip scores3 in the current study are comparable with those that have been reported in earlier studies. Tompkins et al.27 reported a mean score of 90 points (range, 44 to 100 points) at seven to ten years, and Lachiewicz et al.11 reported a mean score of 94 points (range, 60 to 100 points) at two to five years. In the present study, four patients had a Harris hip score of less than 45 points, which could not be explained by the radiographic findings. Like Schmalzried and Harris22, we found gaps between the cup and the bone on the postoperative radiographs that could not be seen on the ten-year radiographs. We attribute this observation to the formation of new bone. The radiolucent lines that were seen at ten years were typically one-half millimeter in thickness or less and were located between a sclerotic line and the surface of the acetabular component. The superiority of screws, compared with pegs or spikes, with regard to the ability to achieve initial stability of acetabular components inserted without cement was demonstrated in a cadaver model10. It is interesting that, in our study of arthroplasties with use of so-called line-to-line reaming of the acetabulum, we did not find that the number of screws that were used had a predictive value; screws probably have no effect on long-term stability20. The Harris-Galante-I cup has eleven screw-holes and was inserted without the so-called press-fitting that is achieved by underreaming of the acetabulum. Currently, recommendations for the insertion of many acetabular cups without cement include one to two millimeters of underreaming to enhance initial stability. Furthermore, few or no screw-holes are recommended to avoid migration of particles, which can promote osteolysis of the acetabular roof. However, we found no evidence of osteolysis in our study. Only one cup had migrated measurably. The estimated precision of measurements of migration is assumed to be approximately two millimeters14. Önsten et al. used roentgen stereophotogrammetric methods for twenty-one patients who had had a Harris-Galante-I cup for more than two years and found no migration of more than two millimeters17. As mentioned, we had problems determining the demarcation between the liner and the ceramic head on radiographs, and therefore only a small number of hips could be analyzed for wear. A wear rate of 0.07 millimeter per year was reported earlier with use of a ceramic femoral head but in association with other prosthetic designs25. This rate seems low; however, comparative studies of acetabular liners with different femoral heads, removed at revision or post mortem, should be performed. We encountered a number of fractured ceramic heads in our department5, some years after the current series, and we therefore reintroduced the use of metallic heads for all hip replacements. Even though about one-third of the patients (113 hips) had died by the time of the most recent follow-up, in the current study of 185 hips that could be evaluated the ten-year results (which included revision of three cups because of aseptic loosening, possible loosening of one cup, and displacement of one cup early after the operation) were good to excellent. This justifies the continued use of this type of cup. However, the problems of revision due to dislocation, fracture of the ceramic head, and dislocation of the liner must still be addressed through further technological advances.
*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. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Zimmer, Denmark A/S.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |