JBJS | The Fate of Stable Cemented Acetabular Components Retained During Revision of a Femoral Component of a Total Hip Arthroplasty*

The Journal of Bone & Joint Surgery, Volume 81, Issue 12

Articles | December 01, 1999

The Fate of Stable Cemented Acetabular Components Retained During Revision of a Femoral Component of a Total Hip Arthroplasty*

RICHARD A. BERGER, M.D.†; LAURA R. QUIGLEY, M.S.†; JOSHUA J. JACOBS, M.D.†; MITCHELL B. SHEINKOP, M.D.†; AARON G. ROSENBERG, M.D.†; JORGE O. GALANTE, M.D.†, CHICAGO, ILLINOIS

View Disclosures and Other Information

Investigation performed at the Department of Orthopaedic Surgery, Rush-Presbyterian-St. Luke's Medical Center, Chicago

The Journal of Bone & Joint Surgery. 1999; 81:1682-91

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Background: The decision as to whether to revise or retain a well fixed cemented acetabular component during revision of a femoral component is especially difficult; the rate of loosening of cemented acetabular components is high, whereas that of porous-coated acetabular components inserted during revision is low. However, removal of a well fixed cemented acetabular component can result in increased operative morbidity and cost and in loss of acetabular bone. Data that can be used to predict the long-term survival of retained well fixed cemented acetabular components are therefore needed.Methods: We studied the five to thirteen-year clinical and radiographic results in a group of twenty-six consecutive patients in whom a well fixed cemented acetabular component had been retained during revision of a femoral component. Typical demographic data on the patients and information about the components were recorded, and the cemented acetabular components were graded as A through F, according to the system of Ranawat et al., at the time of the femoral revision. The average duration of follow-up was 8.4 years (range, 5.0 to 12.7 years). No patient was lost to follow-up.Results: Four acetabular components (15 percent) had progressive radiolucency (at forty-eight, forty-eight, fifty-nine, and seventy-five months after the femoral revision) and were considered radiographically loose despite not being associated with symptoms. All four components were graded as either E or F at the time that they were retained during the femoral revision; radiographic loosening was significantly related to these two grades (p < 0.01). No acetabular component with a grade of A, B, C, or D loosened. The components that loosened had been in vivo for a relatively shorter, as opposed to longer, duration before the femoral revision compared with the components that did not loosen (p < 0.05).Conclusions: Retention of the well fixed cemented acetabular components was associated with good clinical results but with a 15 percent rate of loosening. Revision of a cemented acetabular component solely on the basis of the duration that it was in vivo or whether a previous revision had been done does not appear to be warranted. Our findings suggest that acetabular components with a grade of A, B, C, or D at the time of a femoral revision may be retained, as these components continued to function at the time of the five to thirteen-year follow-up in the current study.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

Aseptic loosening is currently the most common mechanism of failure of cemented total hip prostheses^{2,4,8,15,25,27,29,30}. The rates of failure of cemented acetabular and femoral components are dissimilar, and failure typically occurs at different times^{6-8,18,19,21,27,29,30,32}. This situation leads to two dilemmas: what should be done with a well fixed cemented femoral component during revision of an acetabular component²⁴ and, less commonly, what should be done with a well fixed cemented acetabular component during revision of a femoral component.

If a well fixed cemented acetabular component is to be retained, the condition of the bearing surface and the integrity of the fixation must be satisfactory. While these factors are often easily determined intraoperatively, predicting the long-term survival of a revised component compared with that of a retained component is more difficult.

Whether to revise or retain a well fixed cemented acetabular component is an especially difficult decision because the rate of loosening of cemented acetabular components has been shown to accelerate with time. Mulroy et al. studied the results of 162 total hip arthroplasties and found, at fifteen years, a 56 percent rate of radiographic loosening of acetabular components that had been inserted with a so-called second-generation cementing technique¹⁹. Moreover, a 2 percent rate of revision of the acetabular component at 11.2 years increased to 16 percent at fifteen years^18,19. In contrast to these disappointing results, the rates of loosening of hemispherical porous-coated acetabular implants have been reported to be very low after both primary and revision total hip arthroplasty^{16,17,21,26,28,30,31}.

Most studies that we are aware of have shown that the long-term results after revision are inferior to those after primary total hip arthroplasty^{3,5,12,14,22,23}. Nonetheless, the reported eight-year rate of loosening of acetabular components inserted without cement during revisions is lower than the rates, after similar durations of follow-up, of loosening of acetabular components inserted with cement during primary procedures^28,29,31. This paradox further complicates the decision as to whether to retain a well fixed cemented acetabular component or to revise it to a component inserted without cement, and it points to the need for the results of retaining such components to be studied.

The purpose of this clinical study was to determine the five to thirteen-year clinical and radiographic results associated with well fixed cemented acetabular components that had been retained at the time of a revision of the femoral component in a series of twenty-six consecutive patients.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Department of Orthopaedic Surgery, Rush-Presbyterian-St. Luke's Medical Center, 1725 West Harrison Street, Suite 1063, Chicago, Illinois 60612.

*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.

†Department of Orthopaedic Surgery, Rush-Presbyterian-St. Luke's Medical Center, 1725 West Harrison Street, Suite 1063, Chicago, Illinois 60612.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 1-A through 1-D: Radiographs of a man who had revision of the retained cemented acetabular component, which was loose but not causing symptoms. Fig. 1-A: Radiograph, made when the patient was fifty-six years old, showing a cemented all-polyethylene acetabular component and a cemented femoral component, which had been inserted four years previously at the time of a revision total hip arthroplasty. When this radiograph was made, the femoral component was loose and was causing pain, necessitating revision. The acetabular cement mantle was grade²⁴ F.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-B: Radiograph made after the femoral component was revised to a component inserted without cement and the acetabular component was retained.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-C: Four years after the femoral revision, the acetabular component was unchanged but the femoral component had subsided and was loose. The patient had pain in the thigh but not in the groin. Soon after this radiograph was made, the femoral component was revised again and the acetabular component was noted to be intraoperatively loose and was revised as well.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-D: Radiograph made after the repeat revision of the femoral component to a fully porous-coated component and the revision of the acetabular component to a component inserted without cement.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 2-A, 2-B, and 2-C: Radiographs of a man who had loosening of the retained acetabular component without symptoms. Fig. 2-A: Radiograph, made when the patient was seventy-seven years old, showing a cemented Ti-BAC acetabular component and a cemented femoral component, which had been inserted five years previously at the time of a primary total hip arthroplasty. When this radiograph was made, the femoral component was loose and painful, necessitating a revision. The acetabular cement mantle was grade²⁴ F.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B: Radiograph made four years after revision of the femoral component to another component inserted with cement, with retention of the Ti-BAC acetabular component, showing the acetabular component starting to loosen.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Seven years after the revision of the femoral component, the acetabular component had migrated superiorly and medially but remained asymptomatic.

View Large | Download Slide(.ppt)
Add to My JBJS

Figs. 3-A and 3-B: Radiographs of a man in whom the retained acetabular component remained well fixed at the time of an eleven-year follow-up. Fig. 3-A: Radiograph made after femoral revision, performed when the patient was sixty-two years old, five years after a primary total hip arthroplasty with insertion of an all-polyethylene acetabular component and a femoral component with cement. The femoral component had become loose and painful and was revised to a component inserted without cement. The acetabular component had a grade-A cement mantle²⁴ and was retained at the time of the revision.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B: Eleven years after the revision of the femoral component, the acetabular cement mantle was still intact, the acetabular component was well fixed, and the hip was functioning well. The acetabular cement mantle was grade²⁴ B.

View Larger

TABLE I POSTOPERATIVE DATA ON THE TWENTY-SIX PATIENTS

Category	No. of Patients
Pain
None	11
Slight	10
Mild	3
Moderate	1
Severe	1
Disabling	0
Limp
None	8
Slight	11
Moderate	4
Severe	3
Unable to walk	0
Support
None	12
Cane for long walks	9
Cane full-time	4
2 canes	0
Crutches	1
Unable to walk	0
Distance walked
Unlimited	12
6 blocks	10
2—3 blocks	4
Indoors only	0
Transfer to bed or chair only	0

Add to My JBJS

TABLE I POSTOPERATIVE DATA ON THE TWENTY-SIX PATIENTS

Category	No. of Patients
Pain
None	11
Slight	10
Mild	3
Moderate	1
Severe	1
Disabling	0
Limp
None	8
Slight	11
Moderate	4
Severe	3
Unable to walk	0
Support
None	12
Cane for long walks	9
Cane full-time	4
2 canes	0
Crutches	1
Unable to walk	0
Distance walked
Unlimited	12
6 blocks	10
2—3 blocks	4
Indoors only	0
Transfer to bed or chair only	0

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

Between February 1985 and June 1993, thirty well fixed cemented acetabular components were retained in thirty consecutive patients during revisions of the femoral component that were performed by three of us (A. G. R., J. J. J., and J. O. G.), who are senior-staff joint-replacement surgeons at Rush-Presbyterian-St. Luke's Medical Center. At the time of the femoral revision, the average age of the twenty-one men and the nine women was sixty-five years (range, thirty-one to eighty-four years) and the cemented acetabular components had been in place for an average of seven years (range, three to sixteen years).

No patient was lost to follow-up. Four patients (13 percent) died during the course of follow-up, of causes unrelated to the arthroplasty. At the time of death, all four hips were functioning well with no radiographic evidence of loosening. The remaining twenty-six patients (twenty-six hips) were alive at a minimum of five years postoperatively; therefore, the study group comprised these patients, which included nineteen men and seven women. All twenty-six patients had clinical and corresponding radiographic follow-up for an average of 8.4 years (range, 5.0 to 12.7 years).

The average age of the twenty-six patients at the time that the acetabular component was retained during the femoral revision was sixty-four years (range, thirty-one to eighty-four years). The reason for the femoral revision was radiographic loosening with pain in the thigh in eighteen patients (69 percent) in whom the femoral component had been inserted with cement and in one (4 percent) in whom the femoral component had been inserted without cement, radiographic loosening with progressive bone loss due to osteolysis in four patients (15 percent) in whom the femoral component had been inserted with cement, and radiographic loosening with fracture of the stem in three patients (12 percent) in whom the femoral component had been inserted with cement. Seventeen patients had failure of a femoral component that had been inserted during a primary total hip arthroplasty and nine, failure of a femoral component that had been inserted during a revision total hip arthroplasty. The acetabular component that had been inserted during the primary or revision total hip arthroplasty was retained in all patients.

The decision to retain the cemented acetabular component was based on both clinical and radiographic considerations. First, the component had to be radiographically stable. Radiographic loosening was defined as the presence of a complete radiolucent line of more than one millimeter in width at the bone-cement interface or any progressive migration or tilting of the component²⁹. Second, there had to have been minimum wear as evidenced radiographically. Third, the position and orientation of the component had to be acceptable; the Köhler line had to be intact (no protrusio), the component had to be at the level of a normal hip center (not elevated more than 1.5 centimeters), and the cup had to be in 40 to 60 degrees of abduction without retroversion. Finally, the component could not be associated with a history of dislocation and the final revision construct had to be intraoperatively stable. The size of the femoral head was not a limiting factor. One acetabular component had an inner diameter of twenty-two millimeters; eleven, twenty-eight millimeters; and fourteen, thirty-two millimeters.

All of the femoral revisions were performed in an operating room that was equipped with vertical laminar airflow, and the operating team wore body-exhaust systems. Intraoperative specimens of joint fluid and capsule were taken for culture. Frozen sections from the capsule and the membrane of the femoral canal also were analyzed. Immediately after the specimens were obtained, the patients received one gram of cefazolin intravenously as prophylaxis against infection.

The acetabular component was evaluated intraoperatively in all patients. To assess the intraoperative stability of the component, the hip was first exposed and dislocated. If motion of the acetabular component was not seen during dislocation, all soft tissue was cleared from the acetabular interface so that both the cement-bone and the cement-prosthesis interface could be visualized. An impactor and a hemostat were used to apply stresses manually to the acetabular component. The cement-bone and cement-prosthesis interfaces were inspected for motion or extrusion of blood with use of manual loading and traction, and either finding was considered to be evidence of loosening²⁰. If inspection of the articular surface revealed no wear, pitting, or abrasions and the final construct was stable, the component was retained.

Two independent reviewers graded the cement-bone interface of each of the thirty retained acetabular components on radiographs that had been made before the revision. These reviewers were an attending orthopaedic surgeon (R. A. B.) who had not been involved in the operations and an orthopaedic research associate who interprets all of the radiographs for our adult reconstruction section. The cement mantle of the acetabular component was graded according to the method described by Ranawat et al.²⁴. The cement-bone interface was divided into three equal zones, and the lateral zone was further subdivided into two equal subzones. The cement-bone interface in the three zones was graded with a point system. An assignment of 1 point indicated good interdigitation of the cement into the underlying cancellous bone, with no radiolucency in the zone; 2 points, a thin radiolucent line of less than one millimeter in width; 3 points, a wide radiolucent line of more than one millimeter in width; and 4 points, migration. The point values for each of the three zones (with the values for the two lateral subzones averaged to obtain a point value for the entire lateral zone) were totaled to obtain a composite score. The minimum score (a perfect cement-bone interface) was 3 points, and the maximum score (migration) was 12 points. The final grade of the acetabular component was based on the composite score, with 3 points being equivalent to grade A; 3.5 points, grade B; 4 points, grade C; 4.5 or 5 points, grade D; 5.5 or 6 points, grade E; and more than 6 points, grade F.

At the time of the femoral revision, when the acetabular component was retained, five acetabular components were grade A; five, grade B; five, grade C; nine, grade D; three, grade E; and three, grade F. The grade of the acetabular component was independent of the duration that the component had been in vivo. Of the twenty acetabular components that had been inserted during a primary procedure, two were grade A; four, grade B; three, grade C; eight, grade D; two, grade E; and one, grade F. Of the ten acetabular components that had been inserted during a revision procedure, three were grade A; one, grade B; two, grade C; one, grade D; one, grade E; and two, grade F. Although the average grade of the acetabular components that had been inserted during a primary procedure was better than that of the components that had been inserted during a revision, with the numbers available this difference was not found to be significant (p = 0.28).

Sixteen of the thirty revision femoral components were inserted without cement, and fourteen were inserted with cement. Twelve BIAS components (Zimmer, Warsaw, Indiana), two Harris-Galante porous-coated components (Zimmer), and two fully porous-coated Solution components (DePuy, Warsaw, Indiana) were inserted without cement. Thirteen Harris Precoat components and one Anatomic component (both made by Zimmer) were inserted with cement, with use of modern cementing techniques. The canal was plugged with a polyethylene cement restrictor, lavaged, and filled in a retrograde fashion with Simplex-P cement (Howmedica, Rutherford, New Jersey) with use of a cement-injector gun, and the cement was pressurized with a proximal femoral seal. All femoral components had a Morse neck taper with a corresponding modular head that matched the inner diameter of the retained acetabular component.

In the hips in which cement was not used to fix the femoral component, a long-stem BIAS or Solution component was used to span femoral defects. The length of the implant was chosen to extend beyond the defect for a distance equal to at least two diameters of the femoral shaft. One intraoperative fracture of the proximal part of the femur was noted in this group of sixteen hips. The fracture was repaired with cerclage wires.

All patients received cefazolin (one gram every eight hours) as prophylaxis against infection until the closed suction drain was removed, usually at forty-eight hours. Beginning on the day of the operation, all patients also received low-dose warfarin as prophylaxis against thromboembolism in order to maintain a prothrombin time of 1.5 times the control value for four weeks.

Physical therapy was begun on the first or second postoperative day. Patients in whom the femoral component had been inserted with cement were allowed to bear weight as tolerated with use of two crutches or a walker for six weeks and then as tolerated without use of assistive devices. Patients in whom the femoral component had been inserted without cement were permitted touch-down weight-bearing with use of two crutches or a walker for six weeks followed by weight-bearing as tolerated with use of assistive devices for six weeks and then by weight-bearing as tolerated without use of assistive devices.

The patients were evaluated clinically and radiographs of the hip and pelvis were made for each patient at predetermined intervals of six weeks, three months, six months, and yearly thereafter. Preoperatively and at each yearly interval, the patients were evaluated with use of the Harris hip-scoring system¹⁰ by independent evaluators. An independent observer also evaluated the components on the postoperative radiographs, with use of radiographs made before the revision as a baseline for the evaluation of the acetabular components and those made at six weeks as a baseline for the evaluation of the femoral components.

A digitizing tablet (Sigma Scan; Jandel Scientific, Corte Madera, California) was used for the quantitative radiographic evaluation of the acetabulum. The acetabular angle, the vertical distance of the center of the acetabulum from the interteardrop line, and the wear of the component were measured as described by Martell et al.¹⁶. The radiographs were scaled for magnification according to the known diameter of each femoral head.

A qualitative evaluation of the acetabulum also was conducted. The implant-bone interface was evaluated for the presence and extent of radiolucent lines in each of the three zones described by DeLee and Charnley⁹. The periacetabular bone also was examined for evidence of osteolysis. The component was considered to be definitely loose on the follow-up radiographs if it had migrated more than two millimeters and as probably loose if there was a radiolucent line of at least two millimeters in width in at least two zones.

The cement-bone interface of the femoral stems inserted with cement and the prosthesis-bone interface of those inserted without cement were analyzed for the presence of osteolysis. Subsidence was measured as the distance between lines drawn perpendicular to the long axis of the prosthesis at the level of the tip of the greater trochanter and at the lateral shoulder of the prosthesis. Subsidence of two millimeters or more was considered important¹⁶. Any shift in the angle of the prosthesis was measured by subtending the longitudinal axis of the prosthesis and the longitudinal axis of the proximal aspect of the femoral shaft, with changes of 2 degrees or more considered important¹⁶. Femoral remodeling also was noted.

Loosening of a femoral component inserted with cement was defined with use of the criteria of Harris et al.¹¹, whereas loosening of one inserted without cement was defined with use of the criteria of Martell et al.¹⁶.

Survivorship analysis of the acetabular components was performed with use of the method of Kaplan and Meier¹³. All thirty hips were included in this analysis, which was conducted twice: first with revision as the end point and then with revision or radiographic loosening as the end point. Statistical analysis was conducted with a nonparametric chi-square test, with use of the Yates and Bonferroni corrections where appropriate, and also with analysis of variance.

Results

Introduction | Materials and Methods | Results | Discussion | References

The average Harris hip score¹⁰ for the twenty-six patients who were followed for at least five years was 54 points (range, 40 to 86 points) before the femoral revision compared with 83 points (range, 40 to 100 points) at an average of 8.4 years postoperatively; this increase was significant (p < 0.01). Ten hips (38 percent) had an excellent result (90 to 100 points), seven hips (27 percent) had a good result (80 to 89 points), five hips (19 percent) had a fair result (70 to 79 points), and four hips (15 percent) had a poor result (less than 70 points). Only one poor result was related to the total hip arthroplasty. The remaining three poor results were due to concomitant severe spinal stenosis in two patients and to polyarticular rheumatoid arthritis in one patient.

At the time of the latest follow-up, twenty-one patients had no or slight pain, three had mild pain, one had moderate pain, and one had severe pain (Table I). Nineteen patients had no or a slight limp, four had a moderate limp, and three had a severe limp. No patient was unable to walk.

There were six reoperations during the follow-up period. Two patients had aseptic loosening of the femoral component (a Precoat component in one and a BIAS component in the other) and were managed with a repeat revision at forty-eight and seventy-five months. In both of these patients, the acetabular component also was revised because it was found to be loose intraoperatively (Figs. 1-A, 1-B, 1-C, 1-D). Both patients had had radiographic evidence of loosening of the acetabular component without migration before the repeat femoral revision. Two additional patients had a repeat revision of the femoral component (a Precoat component in one and a BIAS component in the other) because of aseptic loosening without alteration of the acetabular component. There were two additional operations, consisting of open reduction and internal fixation of a postoperative femoral fracture without alteration of the acetabular component, in two of the patients who were followed for less than five years. No patient had a reoperation because of infection or dislocation.

Radiographs showed that one acetabular component had migrated; it had been considered to be aseptically loose at forty-eight months after the femoral revision (Figs. 2-A, 2-B, and 2-C). Three additional acetabular components had progressive radiolucency and were considered to be probably loose at forty-eight, fifty-nine, and seventy-five months. Therefore, a total of four acetabular components were considered to be aseptically loose. Two of these loose components were associated with concurrent loosening of the femoral component. The loosening of the femoral component caused pain, and both components in these two patients subsequently were revised (Figs. 2-A, 2-B, and 2-C). The other two radiographically loose acetabular components were not associated with bone loss or symptoms, and these two patients had a Harris hip score¹⁰ of 94 and 80 points at ninety-one and 128 months. No osteolysis was seen around any stable acetabular component.

None of the four patients who were followed for less than five years had revision or radiographic loosening of the acetabular component. As mentioned, two patients had open reduction and internal fixation of a postoperative femoral fracture without alteration of the acetabular component. None of these patients had a reoperation because of infection or dislocation.

Kaplan-Meier analysis¹³ of all thirty hips showed that the probability of the cemented acetabular component surviving for ten years was 92 percent (95 percent confidence interval, 87 to 97 percent) with revision as the end point and 83 percent (95 percent confidence interval, 75 to 91 percent) with revision or radiographic evidence of aseptic loosening as the end point.

Failure of the cemented acetabular component was significantly related to the grade²⁴ of its cement mantle at the time of the revision of the femoral component (p < 0.01, chi-square test). If A, B, C, and D are considered to be acceptable grades and E and F, to be unacceptable grades, then all four acetabular components that had aseptic loosening had had an unacceptable grade at the time of the femoral revision. No acetabular component with an acceptable grade loosened (Figs. 3-A and 3-B). One of the three grade-E and all three of the grade-F acetabular components were considered to be aseptically loose at the time of follow-up.

None of the fifteen acetabular components that initially had had a grade-A, B, or C cement mantle had deterioration to a grade of E or F on follow-up radiographs, whereas three of the nine acetabular components that initially had had a grade of D had deterioration to grade E. However, none of these three components were considered to be radiographically loose.

The aseptically loose acetabular components tended to have been in vivo for a shorter, not a longer, duration before the femoral revision. The acetabular components that failed had been in vivo for an average of 3.5 years (range, 2.6 to 4.3 years) before the femoral revision, whereas those that did not fail had been in vivo for an average of 7.4 years (range, 2.9 to 15.6 years); this difference was significant (p < 0.05, chi-square test). Three of the four acetabular components that were aseptically loose had been inserted during a revision procedure. However, with the numbers available, the difference between the rate of loosening for components inserted during a primary procedure and the rate for those inserted during a revision procedure was not found to be significant (p = 0.08, chi-square test). Furthermore, when the grade²⁴ of the acetabular component was taken into account, no association was found between these rates (p = 0.85). On the basis of the numbers available, no other factor, including use of a metal backing (p = 0.21), the total duration that the component had been in vivo (p = 0.31), the initial type of stem (p = 0.85), the type of stem used in the revision (p = 0.77), gender (p = 0.94), and age (p = 0.72), was found to be significantly related to aseptic loosening.

Four femoral components (two that had been inserted with cement and two that had been inserted without cement) subsided and were considered to be loose. The two cemented femoral components that had subsided and were associated with pain were revised to a Precoat component inserted with cement. Of the two femoral components inserted without cement that failed, one was revised to a fully porous-coated stem and the other, to a stem inserted with cement. No patient had femoral osteolysis without loosening.

There was a total of five complications in the thirty patients. One patient had an intraoperative femoral fracture, two had a postoperative femoral fracture, one had a peroneal nerve palsy, and one had a presumed infection. The last patient was presumed to have an infection because Staphylococcus epidermidis grew on culture of intraoperative specimens. The infection was treated with intravenous administration of antibiotics for six weeks, and the patient was asymptomatic with both components in place at 128 months. The intraoperative fracture was treated with cerclage wires, and it healed uneventfully. Both postoperative femoral fractures occurred less than five years after the revision: one occurred at two years, when the allograft portion of an allograft-prosthesis composite fractured, and the other occurred at five months, in zone 5 around a cemented component. Both fractures healed after open reduction and internal fixation. The peroneal nerve palsy was an exacerbation of a previous partial peroneal nerve palsy that had been present preoperatively. The palsy worsened at the time of the femoral revision but subsequently returned to the preoperative level without intervention.

There were no dislocations or clinically evident pulmonary embolisms.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

Removal of a well fixed cemented acetabular component during a femoral revision can result in increased operative morbidity and cost and in acetabular bone loss^1,5,22. Retention of such a component at the time of a femoral revision therefore may be contemplated. In deciding whether to retain the component, the surgeon must consider the average longevity of a retained cemented acetabular component compared with that of a revision acetabular component inserted without cement. However, an assumption that the longevity of a retained cemented acetabular component in a hip in which the femoral component has been revised is similar to that of an acetabular component in a hip with a well fixed femoral component that was not revised may lead to an erroneous decision. In an effort to resolve this problem, we investigated the fate of well fixed cemented acetabular components that had been retained during femoral revision.

The present study demonstrated an 8 percent rate of revision of the acetabular component (two of twenty-six) with an additional 8 percent rate of radiographic loosening without revision (two of twenty-six) at the five to thirteen-year follow-up evaluation. To assess the efficacy of retention of cemented acetabular components, this 15 percent rate of loosening (four of twenty-six) must be compared with current rates of failure of revised acetabular components.

Hemispherical porous-coated acetabular implants have been shown to have a substantially lower rate of loosening than cemented acetabular components after both primary and revision procedures^{16,17,21,26,28,30}. Silverton et al. reported on 138 consecutive acetabular revisions that were performed at our institution during an era similar to that of the current study (from 1983 to 1986), with a similar duration of follow-up (average, eight years)²⁸. They found an 11 percent rate of repeat revision (thirteen of 115 hips), with no repeat acetabular revisions because of aseptic loosening. Four hips (3 percent) had a repeat acetabular revision because of recurrent dislocation; six (5 percent), because of septic loosening; and three (3 percent) were done concomitant with a repeat revision of the femoral component. One hundred (92 percent) of 109 acetabular components were determined to be radiographically stable, seven (6 percent) were considered to be possibly unstable, one (1 percent) was thought to be probably unstable, and one (1 percent) was judged to be septically loose²⁸. Therefore, the combined rate of mechanical failure was 1 percent (one of 109).

The poor ten-year rate of survival of the retained cemented acetabular components in the current series compared with the excellent rate of survival of the acetabular components inserted without cement during revisions in the series of Silverton et al.²⁸ seems to suggest that the retention of cemented components is not effective. However, compared with revision of a loose component, revision of a well fixed component is more difficult, necessitates more operative time with increased morbidity and expense, and may be associated with a higher rate of failure^1,5,22,23. Furthermore, none of the patients in the current series needed an additional procedure specifically because of aseptic loosening of the acetabular component. Two radiographically loose acetabular components were revised; however, the indication for the revision was not loosening of the acetabular component but, rather, loosening of the femoral component (that is, the acetabular revision was done incidentally). The two acetabular components that remained radiographically loose were not associated with pain or bone loss, and they were not being considered for revision at the time of this writing. Therefore, retention of the cemented acetabular components resulted in no additional procedures or morbidity at five to thirteen years of follow-up.

The analysis of the failures in this series is valuable with regard to future decisions as to whether to revise or retain well fixed cemented acetabular components. The ultimate failure of the retained cemented acetabular component was found to be significantly related to the grade of the cement mantle of the acetabular component according to the grading scale of Ranawat et al.²⁴ at the time of the femoral revision (p < 0.01, chi-square test). All four of the aseptically loose acetabular components had an unacceptable appearance of the cement (grade E or F) when they were retained at the time of the femoral revision. No acetabular component with a grade of A, B, C, or D loosened during the follow-up period. One of the three grade-E components and all three of the grade-F components had radiographic evidence of loosening at the time of follow-up. These findings are similar to those reported by Ranawat et al., in a study of 502 acetabular components that had been inserted with cement during a primary procedure²⁴. At an average of nine years (range, five to fifteen years) postoperatively, 11 percent (nine) of eighty grade-D acetabular components, two of fourteen grade-E components, and all three grade-F components had failed. Those authors concluded that grades A, B, and C were acceptable, whereas grades D, E, and F were unacceptable. In the current study, failure of the cemented acetabular component was significantly related to the grade of the cementing technique, with the grade assigned with use of the scale of Ranawat et al.²⁴ at the time of a femoral revision predictive of failure of the acetabular component. However, our findings suggest that only grades of E and F are unacceptable, as only those grades were associated with loosening. Although no grade-D acetabular component failed, three of nine such components had deterioration to a grade of E at the latest follow-up evaluation. This deterioration may be ominous with regard to the long-term survival of grade-D acetabular components. Therefore, caution may be appropriate when retention of a grade-D component is being contemplated.

Contrary to previous findings that cemented acetabular components with a longer duration in vivo have an increased rate of failure^{19,27,29,31,33}, the results of our study do not support revision of a well fixed cemented acetabular component on the basis of a long duration in vivo alone. In fact, we found the converse to be true: the acetabular components that failed had been in vivo for an average of 3.5 years at the time of the femoral revision, whereas those that did not fail had been in vivo for an average of 7.4 years. This suggests that acetabular components that withstood a longer duration in vivo and still maintained a good cement mantle were less likely to fail with continued use. Similarly, if a good cement mantle (grade A, B, C, or D) had been maintained, an acetabular component that had been inserted during a revision was not more likely to fail than one that had been inserted during a primary procedure. None of the retained acetabular components that had been inserted during a revision and that had grade-A, B, C, or D cement loosened during the follow-up period.

In conclusion, retention of well fixed cemented acetabular components yielded good clinical results but a 15 percent rate of radiographic failure. Revision of a cemented acetabular component solely because it had been in vivo for a long duration or had been inserted during a revision does not appear to be warranted. A careful radiographic evaluation with use of the grading system of Ranawat et al.²⁴ should be performed before it is decided whether to retain a cemented acetabular component. Acetabular components with grade-E or F cement should be revised. We believe that acetabular components with grade-A, B, C, or D cement may be retained if the position, orientation, and degree of wear are acceptable, even though some grade-D components in the current series had deterioration to grade E without loosening at the time of the latest follow-up. These conclusions are based on the finding that no acetabular component with grade-A, B, C, or D cement failed during the five to thirteen-year follow-up period.

References

Introduction | Materials and Methods | Results | Discussion | References

Ahnfelt, L.; Herberts, P.; Mulchau, H.; and Andersson, G. B.: Prognosis of total hip replacement. A Swedish multicenter study of 4,664 revisions. Acta Orthop. Scandinavica, Supplementum 238, 1990.

Amstutz, H. C.; Markolf, K. L.; McNeice, G. M.; and Gruen, T. A.: Loosening of total hip components: cause and prevention. In The Hip. Proceedings of the Fourth Open Scientific Meeting of the Hip Society, pp. 102-116. St. Louis, C. V. Mosby, 1976.

Amstutz, H. C.; Ma, S. M.; Jinnah, R. H.; and Mai, L.: Revision of aseptic loose total hip arthroplasties. Clin. Orthop.,170: 21-33, 1982.17021 1982 [PubMed]

Beckenbaugh, R. D., and Ilstrup, D. M.: Total hip arthroplasty. A review of three hundred and thirty-three cases with long follow-up. J. Bone and Joint Surg.,60-A: 306-313, April 1978.60-A306 1978

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

Charnley, J.: Anchorage of the femoral head prosthesis to the shaft of the femur. J. Bone and Joint Surg.,42-B(1): 28-30, 1960.42-B(1)28 1960

Charnley, J.: Fracture of femoral prostheses in total hip replacement. A clinical study. Clin. Orthop.,111: 105-120, 1975.111105 1975 [PubMed]

Charnley, J.: Low Friction Arthroplasty of the Hip. Theory and Practice, p. 64. New York, Springer, 1979.

DeLee, J. G., and Charnley, J.: Radiological demarcation of cemented sockets in total hip replacement. Clin. Orthop.,121: 20-32, 1976.12120 1976 [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

Harris, W. H.; McCarthy, J. C., Jr.; and O'Neill, D. A.: Femoral component loosening using contemporary techniques of femoral cement fixation. J. Bone and Joint Surg.,64-A: 1063-1067, Sept. 1982.64-A1063 1982

Hunter, G. A.; Welsh, R. P.; Cameron, H. U.; and Bailey, W. H.: The results of revision of total hip arthroplasty. J. Bone and Joint Surg.,61-B(4): 419-421, 1979.61-B(4)419 1979

Kaplan, E. L., and Meier, P.: Nonparametric estimation from incomplete observations. J. Am. Statist. Assn.,53: 457-481, 1958.53457 1958

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

Ling, R. S. M.: Loosening experiences at Exeter. Orthop. Trans.,5: 351, 1981.5351 1981

Martell, J. M.; Pierson, R. H., III; Jacobs, J. J.; Rosenberg, A. G.; Maley, M.; and Galante, J. O.: Primary total hip reconstruction with a titanium fiber-coated prosthesis inserted without cement. J. Bone and Joint Surg.,75-A: 554-571, April 1993.75-A554 1993

Mohler, C. G.; Kull, L. R.; Martell, J. M.; Rosenberg, A. G.; and Galante, J. O.: Total hip replacement with insertion of an acetabular component without cement and a femoral component with cement. Four to seven-year results. J. Bone and Joint Surg.,77-A: 86-96, Jan. 1995.77-A86 1995

Mulroy, R. D., Jr., and Harris, W. H.: The effect of improved cementing techniques on component loosening in total hip replacement. An 11-year radiographic review. J. Bone and Joint Surg.,72-B(5): 757-760, 1990.72-B(5)757 1990

Mulroy, W. F.; Estok, D. M.; and Harris, W. H.: Total hip arthroplasty with use of so-called second-generation cementing techniques. A fifteen-year-average follow-up study. J. Bone and Joint Surg.,77-A: 1845-1852, Dec. 1995.77-A1845 1995

O'Neill, D. A., and Harris, W. H.: Failed total hip replacement: assessment by plain radiographs, arthrograms, and aspiration of the hip joint. J. Bone and Joint Surg.,66-A: 540-546, April 1984.66-A540 1984

Padgett, D. E.; Kull, L.; Rosenberg, A.; Sumner, D. R.; and Galante, J. O.: Revision of the acetabular component without cement after total hip arthroplasty. Three to six-year follow-up. J. Bone and Joint Surg.,75-A: 663-673, May 1993.75-A663 1993

Pellicci, P. M.; Wilson, P. D., Jr.; Sledge, C. B.; Salvati, E. A.; Ranawat, C. S.; Poss, R.; and Callaghan, J. J.: Long-term results of revision total hip replacement. A follow-up report. J. Bone and Joint Surg.,67-A: 513-516, April 1985.67-A513 1985

Peters, C. L.; Kull, L.; Jacobs, J. J.; Rosenberg, A. G.; and Galante, J. O.: The fate of well fixed cemented femoral components left in place at the time of revision of the acetabular component. J. Bone and Joint Surg.,79-A: 701-706, May 1997.79-A701 1997

Ranawat, C. S.; Deshmukh, R. G.; Peters, L. E.; and Umlas, M. E.: Prediction of the long-term durability of all-polyethylene cemented sockets. Clin. Orthop.,317: 89-105, 1995.31789 1995 [PubMed]

Salvati, E. A.; Wilson, P. D., Jr.; Jolley, M. N.; Vakili, F.; Aglietti, P.; and Brown, G. C.: A ten-year follow-up study of our first one hundred consecutive Charnley total hip replacements. J. Bone and Joint Surg.,63-A: 753-767, June 1981.63-A753 1981

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

Schulte, K. R.; Callaghan, J. J.; Kelley, S. S.; and Johnston, R. C.: The outcome of Charnley total hip arthroplasty with cement after a minimum twenty-year follow-up. The results of one surgeon. J. Bone and Joint Surg.,75-A: 961-975, July 1993.75-A961 1993

Silverton, C. D.; Rosenberg, A. G.; Sheinkop, M. B.; Kull, L. R.; and Galante, J. O.: Revision total hip arthroplasty using a cementless acetabular component. Technique and results. Clin. Orthop.,319: 201-208, 1995.319201 1995 [PubMed]

Stauffer, R. N.: Ten-year follow-up study of total hip replacement, with particular reference to roentgenographic loosening of the components. J. Bone and Joint Surg.,64-A: 983-990, Sept. 1982.64-A983 1982

Sumner, 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

Sutherland, C. J.; Wilde, A. H.; Borden, L. S.; and Marks, K. E.: A ten-year follow-up of one hundred consecutive Müller curved-stem total hip-replacement arthroplasties. J. Bone and Joint Surg.,64-A: 970-982, Sept. 1982.64-A970 1982

Tompkins, G. S.; Jacobs, J. J.; Kull, L. R.; Rosenberg, A. G.; and Galante, J. O.: Primary total hip arthroplasty with a porous-coated acetabular component. Seven-to-ten-year results. J. Bone and Joint Surg.,79-A: 169-176, Feb. 1997.79-A169 1997

Wroblewski, B. M.: 15-21-year results of the Charnley low-friction arthroplasty. Clin. Orthop.,211: 30-35, 1986.21130 1986 [PubMed]

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-B: Radiograph made after the femoral component was revised to a component inserted without cement and the acetabular component was retained.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-D: Radiograph made after the repeat revision of the femoral component to a fully porous-coated component and the revision of the acetabular component to a component inserted without cement.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-C Seven years after the revision of the femoral component, the acetabular component had migrated superiorly and medially but remained asymptomatic.

View Large | Download Slide(.ppt)
Add to My JBJS

View Larger

TABLE I POSTOPERATIVE DATA ON THE TWENTY-SIX PATIENTS

Category	No. of Patients
Pain
None	11
Slight	10
Mild	3
Moderate	1
Severe	1
Disabling	0
Limp
None	8
Slight	11
Moderate	4
Severe	3
Unable to walk	0
Support
None	12
Cane for long walks	9
Cane full-time	4
2 canes	0
Crutches	1
Unable to walk	0
Distance walked
Unlimited	12
6 blocks	10
2—3 blocks	4
Indoors only	0
Transfer to bed or chair only	0

Add to My JBJS

TABLE I POSTOPERATIVE DATA ON THE TWENTY-SIX PATIENTS

Category	No. of Patients
Pain
None	11
Slight	10
Mild	3
Moderate	1
Severe	1
Disabling	0
Limp
None	8
Slight	11
Moderate	4
Severe	3
Unable to walk	0
Support
None	12
Cane for long walks	9
Cane full-time	4
2 canes	0
Crutches	1
Unable to walk	0
Distance walked
Unlimited	12
6 blocks	10
2—3 blocks	4
Indoors only	0
Transfer to bed or chair only	0