- Email: [email protected]
Fifteen-Year to Twenty-Year Results of Cementless Harris-Galante Porous Femoral and Harris-Galante Porous I and II Acetabular Components
The Journal of Arthroplasty Vol. 25 No. 5 2010
Fifteen-Year to Twenty-Year Results of Cementless Harris-Galante Porous Femoral and Harris-Galante Porous I and II Acetabular Components Scott D. Anseth, MD,* Pamela A. Pulido, BSN,y Wendy S. Adelson, MS,y Shantanu Patil, MD,y Julie C. Sandwell, MPH(c),y and Clifford W. Colwell Jr, MDy
Abstract: This study examined long-term survivorship of primary total hip arthroplasty (THA) using cementless Harris-Galante porous femoral and Harris-Galante porous I or II acetabular components. Of 113 hips (101 patients) studied, 60 hips (53 patients) were available for follow-up at a mean of 17.2 years after surgery. A total of 10 hips had documented revision, and 2 hips had failed radiographically. The average Harris hip score was 82. Radiographically, 12 hips demonstrated eccentric wear, 8 had osteolysis, and 1 had a broken tine. The overall survival rate was 87.7%; the mean volumetric wear rate was 74.96 mm3; and the mean polyethylene linear penetration rate was 0.153 mm/y, similar to that of well-cemented THA in other series. This long-term outcome for an early-generation cementless THA is promising and provides a standard by which to judge the newer generation of cementless implants. Keywords: cementless total hip arthroplasty, revision, femoral head penetration, polyethylene wear, survival. © 2010 Published by Elsevier Inc.
The goals of total hip arthroplasty (THA) include pain relief with rigid fixation, long-term biocompatibility, and favorable remodeling characteristics within the bone [1]. Compared with cemented THA [2,3], the long-term results of early-generation cementless THA have not been as convincing [4]. The success of cementless fixation has been a subject of study since its introduction, and designs have been modified over the years [1,2]. The fundamental theoretical advantage of the cementless approach in THA is achieving adequate bony fixation without the use of cement [5]. This prospective study reports the 15- to 20-year followup of the Harris-Galante Porous (HGP) implant (Zimmer Inc, Warsaw, Ind) THA performed by a single surgeon examining the long-term survivorship and reasons for revision arthroplasty.
From the *Twin Cities Orthopaedics, Minneapolis, Minnesota; and yShiley Center for Orthopaedic Research and Education at Scripps Clinic, Scripps Research Services, La Jolla, California. Submitted November 2, 2008; accepted May 24, 2009. No benefits or funds were received in support of the study. Reprint requests: Clifford W. Colwell, Jr, MD, Shiley Center for Orthopaedic Research and Education at Scripps Clinic, 11025 North Torrey Pines Road, Suite 140, La Jolla, CA 92037. © 2010 Published by Elsevier Inc. 0883-5403/2505-0003$36.00/0 doi:10.1016/j.arth.2009.05.033
Materials and Methods Using our clinical outcomes database (PATS, Axis Clinical Software Inc, Portland, Ore) with institutional review board approval, patients who underwent THA by the senior author between 1985 and 1989 using HGP femoral components with either HGP I or HGP II acetabular components (Fig. 1) were identified. Selection criteria for cementless arthroplasty were advanced hip pathology, relatively younger age, and anthropometric compatibility between the proximal femur and femoral component. The HGP femoral component was an early-generation, noncemented porous ingrowth design made from forged titanium alloy with commercially pure titanium fiber-mesh porous ingrowth pads, the mesh having a pore volume of 50% with an average pore size of 300 μm. The HGP I cup was a press-fit, with multiple screw holes that used 4 tines to lock the modular polyethylene liner into place. The HGP II cup was modified to include additional locking tines. A posterolateral surgical approach with intact trochanter and a 28-mm head were used in all cases. Postoperative rehabilitation was standardized and included non– weight-bearing ambulation for 6 weeks, followed by a gradual increase in weight-bearing for 6 weeks. Full weight-bearing was begun at 3 months if the patient was pain-free and had excellent radiographic evaluation. Stair climbing using the operated limb was allowed after 3 months.
687
688 The Journal of Arthroplasty Vol. 25 No. 5 August 2010 Table 1. Patient Demographics and Preoperative Diagnoses Per Hip at Time of Index Surgery Demographics
No.
Mean
Range
Age, y Height, m Weight, kg
113 109 109
54 1.7 77.1
28-74 1.5-1.9 44.9-117.9
Preoperative Diagnosis
No.
%
Primary osteoarthritis Osteonecrosis Posttraumatic arthritis Other
74 18 6 13
65 16 1 13
Results Fig. 1. Harris-Galante porous II acetabular components and stem.
Follow-up methods used in contacting patients were modeled after those used and well-described by King et al [3]. Families of deceased patients were contacted to find out if the patient had a revision before death. Patients returned for clinical follow-up at 3 months, at 1 year, and at 5-year intervals after surgery. Patients who had not undergone revision surgery returned for clinical follow-up 15 to 20 years after surgery. Follow-up included a physical examination with clinical and radiographic evaluation. Harris hip scores (HHSs) were collected preoperatively and at follow-up, with results classified as excellent (90-100), good (80-89), fair (7079), or poor (b70). Patients who were unable to return for a physical examination were administered the function portion of the HHS over the telephone. Radiographs of the operative hip taken at 3 months and at 15 to 20 years postoperatively in the anteroposterior and lateral views and were evaluated for evidence of implant failure. Specific radiographic evaluation included stem osteolysis, pedestal formation, implant-bone interface evaluation, and polyethylene wear. All radiographs were digitally scanned (Epson Expression 1680, Epson America, Inc, Long Beach, Calif). Radiographic analyses consisting of cup inclination angle, cup anteversion angle, linear penetration rate, and volumetric wear rate were measured using a commercially available software program (Polyware Pro 3D, Draftware Developers Inc, Vevay, Ind). All statistical analyses were conducted using SPSS (Version 13.0 for Windows, SPSS Inc, Chicago, Ill). Radiographic analyses were correlated using the Pearson product-moment correlation. A P value of less than .05 was considered significant. Implant survivorship was characterized using Kaplan-Meier analysis with revision for any reason and for specific reasons as end points. Implant success was defined as the prosthesis remaining in situ.
One hundred one patients (51 males, 50 females, 113 hips) comprised the index group. The distributions of demographic characteristics and preoperative diagnoses are displayed in Table 1. Of the 113 hips, the HGP I cup was used in 85 and the HGP II cup in 28. At follow-up, 32 patients (34 hips) were found to be deceased; 6 patients (9 hips) were lost to follow-up; 10 living patients (10 hips) had undergone revision surgery; and 53 patients (60 unrevised hips) returned for clinical evaluation at a mean of 17.2 years (range, 15-20 years). Of the 32 deceased patients, 11 of 34 hips were known to be unrevised at time of death, but revision status on the remaining 23 hips was unable to be ascertained. Revision arthroplasty on the 10 patients included 8 patients (9.5%) who received HGP I cups and 2 patients (6.9%) who received HGP II cups. The length of survival and reasons for revision surgery are displayed in Fig. 2. The average time to revision surgery was 10.74 years (range, 2.9217.75 years). The HHSs are displayed in Table 2. The function category of HHS is displayed separately to include patients who were not able to complete the entire scoring system. Radiographically, 12 hips demonstrated eccentric wear; 6 had femoral osteolysis (Fig. 3); 2 had acetabular osteolysis; and 1 had a broken tine (Fig. 4). A subgroup assessment using 2-dimensional analysis performed on 25 of the hips for which both baseline (3 months
Fig. 2. Time course and reasons for revision surgery during 15to 20-year follow-up.
HGP Femoral and HGP I and II Acetabular Components Anseth et al
689
Table 2. Harris Hip Scores and Clinical Outcome of Unrevised THA Patients at 15 to 20 Years No. HHS (0–100) Preoperative 49 At follow-up 39 Harris hip function score (0-47) At follow-up 48
Mean ± SD
Range
57 ± 15.5 82 ± 20.0
23-95 30-100
37 ± 9.5
11-47
Clinical outcome
No.
%
Excellent/Good Fair/Poor
25 14
64 36
The Harris hip function score (HHFS) is a subset of the HHS that can be administered by telephone. The 48 patients completing the follow-up HHFS include the 39 who were able to return for follow-up and completed the entire HHS plus 9 patients who could only complete the HHFS section by telephone.
postoperative) and 15- to 20-year follow-up radiographs were available demonstrated a mean linear polyethylene penetration rate of 0.153 mm/y and a mean volumetric
Fig. 4. X-ray depicting a broken tine and asymmetric poly wear in the HGP acetabular component.
wear rate of 74.96 mm3 (Table 3). More hips were not evaluated for analysis because of the lack of precision and quality of x-rays, which are necessary to do this measurement. A small difference in polyethylene wear rates between the 2 cups was present with 0.161 mm/y in the HGP I cup and 0.139 mm/y in the HPG II cup. No correlation was found between linear penetration rates and cup anteversion angle or cup abduction angle. Overall, survival status was established for 73 patients representing 81 of the 113 index hips. The survival rate (percentage of hips not requiring a revision) in these patients was 87.7% (71 of 81 were unrevised) at 20 years. Worst-case scenario calculation, with patients either lost to follow-up or of unknown revision status at time of death assumed to be revised, yielded a survival rate of 62.8% (71 of 113 unrevised). Best-case scenario calculation, with patients either lost to follow-up or of unknown revision status at time of death assumed to be unrevised, yielded a survival rate of 91.2% (103 of 113 unrevised). Using the Kaplan-Meier method to calculate implant survivorship, the probability of retaining the implant at
Table 3. Results of Radiographic Analyses Showing Linear Penetration Rate, Volumetric Wear Rate, and Cup Angle at 15 to 20 years
Fig. 3. X-ray demonstrating osteolysis around the femoral stem.
Radiographic Analysis
No.
Linear penetration rate Volumetric wear rate Cup abduction angle Cup anteversion angle
25 25 25 25
Mean ± SD 0.153 ± 74.96 ± 35.54° ± 20.26° ±
0.124 mm/y 75.01 mm3 9.54° 9.31°
690 The Journal of Arthroplasty Vol. 25 No. 5 August 2010
Fig. 5. Survivorship curves showing the probability of implant retention with all-cause and cause-specific revision as the end points.
20 year follow-up with revision for any reason as the end point was 84.7% (95% confidence interval [CI], 74.7%91.5%). The survivorship probabilities for acetabular aseptic loosening (n = 2), polyethylene wear (n = 2), and femoral aseptic loosening (n = 6) as end points were 97.4% (95% CI, 90.4%-99.5%), 96.7% (95% CI, 89.3%99.2%), and 90% (95% CI, 80.8%-95.2%), respectively (Fig. 5).
Discussion The longevity of the HGP acetabular components has been excellent in short- and long-term follow-up studies with regard to resistance to mechanical loosening. Our study documented outstanding survival of the HGP cup with only 2 failures due to acetabular aseptic loosening. Survivorship probabilities were 97.4% for acetabular aseptic loosening revisions and 96.7% for polyethylene wear revisions as end points, similar to what has been reported. One study documented no revision of acetabular components at 15 years because of aseptic loosening [4]. Another study documented 1 revision for aseptic loosening at 6 years [5], and 3 other studies documented no revisions of HGP acetabular components at 10 years for aseptic loosening, but 2 for osteolysis around the cup [2,6,7]. The HGP acetabular component's original weakness was the locking mechanism for securing the polyethylene liner. Several authors reported disassociation of liners from securely fixed acetabular cups [8-10]. Even in the long-term follow-up studies, polyethylene disassociation is a more common reason for revision of the acetabular component than aseptic loosening. Clohisy and Harris [7] reported in 2 different studies that acetabular revision due to liner dislodgement and tine fracture was the reason for 3 of 4 revisions in 1 report and 3 of 8 revisions in the other [6]. Parvizi et al [4] documented in their study that 2 HGP acetabular components were revised at 2 to 3 years because of acetabular liner disassociation. In 1 case, the cup was retained, but in the other, the cup also had to be revised. Tompkins et al [11] reported 3 revisions in a series of HGP hips, 2 for polyethylene exchange and 1
for “massive” retroacetabular osteolysis. In our study we did not find any clinically apparent liner disassociation and had only one radiographically visible broken tine. Several investigations have shown polyethylene wear rates from 0.015 to 0.18 mm/y with the HGP cups implanted during the same period [2,5,11-17]. Some investigations have found that as the wear rate increases, the HGP stem and cups become more susceptible to osteolysis and aseptic failure. D'Lima et al [5], following a cohort with a subset of patients also included in this study, were able to demonstrate a link between polyethylene wear and femoral osteolysis; a positive correlation between the 2 was detectable at a mean follow-up of 6 years, with polyethylene wear at 0.11 mm/y in hips with no femoral osteolysis and at 0.18 mm/y in hips with femoral osteolysis. Two other studies [2,16] documented greater osteolysis surrounding the HGP cups in which the polyethylene wear rate was 0.16 mm/y. Our study demonstrates good long-term survival with a mean polyethylene linear penetration rate of 0.153 mm/y. The difference in mean polyethylene linear penetration between HGP I and HGP II cups was 0.022 mm/y. The fact that a small number of x-rays were evaluated in this manner leaves this study open to a type II statistical error and, therefore, we did not present statistical differences. The other shortcoming of the early-generation HGP cementless THA was the femoral stem, a titanium alloy stem with commercially pure titanium fiber-mesh porous-coating ingrowth pads located on the anterior, the posterior, and the medial surfaces of the proximal third of the stem. These pads provide points of attachment for bony ingrowth, and the smooth channels between the pads provide access for wear particles to affect the femoral shaft. In our study, 6 hips were revised for femoral aseptic loosening before the follow-up period, and 2 hips had radiographically loose femoral components at the time of follow-up. The durability of the HGP femoral component was a concern even in early studies. One study [18] reported femoral lysis was present in 31% of the femurs at 53 months, and another study [15] found osteolysis in 21% of femurs at a mean follow-up of 6 years. Two studies [17,19] reported inferior survivorship of the HGP femoral component with an 86% survivorship at 10 years and a 76.3% survivorship at 13 years, respectively. Parvizi et al [4] reported 8 femoral revisions for aseptic loosening at a mean of 8.2 years and predicted femoral survival would be 82% at 15 years with mechanical failure as the end point, which is comparable to the 84.7% survivorship in our study. The Kaplan-Meier survivorship probability with femoral revision as the end point was 90% (95% CI, 80.8%-95.2%) in our cohort. Petersilge et al [13] in their prospective study, which includes a subset of patients also in this study, documented a postoperative mean HHS of 92 at 5.8 years in a population of cementless HGP hip arthroplasties. Archibeck et al [2] also found a high postoperative
HGP Femoral and HGP I and II Acetabular Components Anseth et al
HHS with second-generation cementless hips, reporting a mean of 94 at 10 years. Kim [20] and Ritter et al [21] documented a steady decrease in HHS over the postoperative period in cementless hips. When considering the known deterioration in HHS with time, our average HHS of 82 at a mean follow-up of 17.2 years is consistent with other long-term studies. As with many long-term longitudinal studies out to 20 years, a number of the patients were lost to follow-up. This was due either to patient death, patients' inability to return for health reasons, or an inability to locate patients. Locating and contacting patients this many years following the index surgery are challenges attested to by the literature [18]. The subgroup that was used to determine penetration wear rate was a small percentage of the entire study group and could have introduced a type 2 error. In our study, the worst-case survival of the HGP hip at 20 years with revision for any reason as the end point was 62.8%, which compares favorably to the 25-year results documented as 63% survival in the living Iowa cemented THA population [22]. Given some of the now apparent design flaws in this early generation of cementless prostheses, with regard to the acetabular locking mechanism and the femoral ingrowth surface, we can reasonably expect the next generation of cementless prostheses to surpass these survivorship numbers when they reach the same period of follow-up.
8.
9.
10.
11.
12.
13.
14.
15.
16.
References 1. Clohisy JC, Harris WH. The Harris-Galante porous-coated acetabular component with screw fixation. An average tenyear follow-up study. J Bone Joint Surg Am 1999;81:66. 2. Archibeck MJ, Berger RA, Jacobs JJ, et al. Secondgeneration cementless total hip arthroplasty. Eight to eleven-year results. J Bone Joint Surg Am 2001;83-A:1666. 3. Rothman RH, Cohn JC. Cemented versus cementless total hip arthroplasty. A critical review. Clin Orthop Relat Res 1990;254:153. 4. Parvizi J, Sullivan T, Duffy G, et al. Fifteen-year clinical survivorship of Harris-Galante total hip arthroplasty. J Arthroplasty 2004;19:672. 5. D'Lima DD, Yashar AA, Venn-Watson EJ, et al. The HarrisGalante porous acetabular component at intermediate follow-up. Orthopedics 2001;24:747. 6. Clohisy JC, Harris WH. The Harris-Galante uncemented femoral component in primary total hip replacement at 10 years. J Arthroplasty 1999;14:915. 7. Clohisy JC, Harris WH. Primary hybrid total hip replacement, performed with insertion of the acetabular component without cement and a precoat femoral component
17.
18.
19.
20.
21.
22.
691
with cement. An average ten-year follow-up study. J Bone Joint Surg Am 1999;81:247. Diekerhof CH, Van Der Hart CP, Fidler MF. Dislocation of polyethylene liner in uncemented Harris Galante II acetabulum: report of 6 cases. J Arthroplasty 2004;19:661. Star MJ, Colwell Jr CW, Donaldson III WF, et al. Dissociation of modular hip arthroplasty components after dislocation. A report of three cases at differing dissociation levels. Clin Orthop Relat Res 1992;111. Gonzalez dV, Ruzo PS, Li S, et al. Dislodgment of polyethylene liners in first and second-generation HarrisGalante acetabular components. A report of eighteen cases. J Bone Joint Surg Am 2001;83-A:553. Tompkins GS, Jacobs JJ, Kull LR, et al. Primary total hip arthroplasty with a porous-coated acetabular component. Seven-to-ten-year results. J Bone Joint Surg Am 1997;79: 169. Cannestra VP, Berger RA, Quigley LR, et al. Hybrid total hip arthroplasty with a precoated offset stem. Four to nine-year results. J Bone Joint Surg Am 2000;82:1291. Petersilge WJ, D'Lima DD, Walker RH, et al. Prospective study of 100 consecutive Harris-Galante porous total hip arthroplasties. 4- to 8-year follow-up study. J Arthroplasty 1997;12:185. Brown III EC, Lachiewicz PF. Precoated femoral component in total hip arthroplasty. Results of 5- to 9-year followup. Clin Orthop Relat Res 1999;153. Latimer HA, Lachiewicz PF. Porous-coated acetabular components with screw fixation. Five to ten-year results. J Bone Joint Surg Am 1996;78:975. Soto MO, Rodriguez JA, Ranawat CS. Clinical and radiographic evaluation of the Harris-Galante cup: incidence of wear and osteolysis at 7 to 9 years follow-up. J Arthroplasty 2000;15:139. Cruz-Pardos A, Garcia-Cimbrelo E. The Harris-Galante total hip arthroplasty: a minimum 8-year follow-up study. J Arthroplasty 2001;16:586. Smith E, Harris WH. Increasing prevalence of femoral lysis in cementless total hip arthroplasty. J Arthroplasty 1995; 10:407. Thanner J, Karrholm J, Malchau H, et al. Poor outcome of the PCA and Harris-Galante hip prostheses. Randomized study of 171 arthroplasties with 9-year follow-up. Acta Orthop Scand 1999;70:155. Kim YH. Long-term results of the cementless porous-coated anatomic total hip prosthesis. J Bone Joint Surg Br 2005;87: 623. Ritter MA, Thong AE, Davis KE, et al. Long-term deterioration of joint evaluation scores. J Bone Joint Surg Br 2004;86:438. Keener JD, Callaghan JJ, Goetz DD, et al. Twenty-five-year results after Charnley total hip arthroplasty in patients less than fifty years old: a concise follow-up of a previous report. J Bone Joint Surg Am 2003;85-A:1066.
Fifteen-Year to Twenty-Year Results of Cementless Harris-Galante Porous Femoral and Harris-Galante Porous I and II Acetabular Components Scott D. Anseth, MD,* Pamela A. Pulido, BSN,y Wendy S. Adelson, MS,y Shantanu Patil, MD,y Julie C. Sandwell, MPH(c),y and Clifford W. Colwell Jr, MDy
Abstract: This study examined long-term survivorship of primary total hip arthroplasty (THA) using cementless Harris-Galante porous femoral and Harris-Galante porous I or II acetabular components. Of 113 hips (101 patients) studied, 60 hips (53 patients) were available for follow-up at a mean of 17.2 years after surgery. A total of 10 hips had documented revision, and 2 hips had failed radiographically. The average Harris hip score was 82. Radiographically, 12 hips demonstrated eccentric wear, 8 had osteolysis, and 1 had a broken tine. The overall survival rate was 87.7%; the mean volumetric wear rate was 74.96 mm3; and the mean polyethylene linear penetration rate was 0.153 mm/y, similar to that of well-cemented THA in other series. This long-term outcome for an early-generation cementless THA is promising and provides a standard by which to judge the newer generation of cementless implants. Keywords: cementless total hip arthroplasty, revision, femoral head penetration, polyethylene wear, survival. © 2010 Published by Elsevier Inc.
The goals of total hip arthroplasty (THA) include pain relief with rigid fixation, long-term biocompatibility, and favorable remodeling characteristics within the bone [1]. Compared with cemented THA [2,3], the long-term results of early-generation cementless THA have not been as convincing [4]. The success of cementless fixation has been a subject of study since its introduction, and designs have been modified over the years [1,2]. The fundamental theoretical advantage of the cementless approach in THA is achieving adequate bony fixation without the use of cement [5]. This prospective study reports the 15- to 20-year followup of the Harris-Galante Porous (HGP) implant (Zimmer Inc, Warsaw, Ind) THA performed by a single surgeon examining the long-term survivorship and reasons for revision arthroplasty.
From the *Twin Cities Orthopaedics, Minneapolis, Minnesota; and yShiley Center for Orthopaedic Research and Education at Scripps Clinic, Scripps Research Services, La Jolla, California. Submitted November 2, 2008; accepted May 24, 2009. No benefits or funds were received in support of the study. Reprint requests: Clifford W. Colwell, Jr, MD, Shiley Center for Orthopaedic Research and Education at Scripps Clinic, 11025 North Torrey Pines Road, Suite 140, La Jolla, CA 92037. © 2010 Published by Elsevier Inc. 0883-5403/2505-0003$36.00/0 doi:10.1016/j.arth.2009.05.033
Materials and Methods Using our clinical outcomes database (PATS, Axis Clinical Software Inc, Portland, Ore) with institutional review board approval, patients who underwent THA by the senior author between 1985 and 1989 using HGP femoral components with either HGP I or HGP II acetabular components (Fig. 1) were identified. Selection criteria for cementless arthroplasty were advanced hip pathology, relatively younger age, and anthropometric compatibility between the proximal femur and femoral component. The HGP femoral component was an early-generation, noncemented porous ingrowth design made from forged titanium alloy with commercially pure titanium fiber-mesh porous ingrowth pads, the mesh having a pore volume of 50% with an average pore size of 300 μm. The HGP I cup was a press-fit, with multiple screw holes that used 4 tines to lock the modular polyethylene liner into place. The HGP II cup was modified to include additional locking tines. A posterolateral surgical approach with intact trochanter and a 28-mm head were used in all cases. Postoperative rehabilitation was standardized and included non– weight-bearing ambulation for 6 weeks, followed by a gradual increase in weight-bearing for 6 weeks. Full weight-bearing was begun at 3 months if the patient was pain-free and had excellent radiographic evaluation. Stair climbing using the operated limb was allowed after 3 months.
687
688 The Journal of Arthroplasty Vol. 25 No. 5 August 2010 Table 1. Patient Demographics and Preoperative Diagnoses Per Hip at Time of Index Surgery Demographics
No.
Mean
Range
Age, y Height, m Weight, kg
113 109 109
54 1.7 77.1
28-74 1.5-1.9 44.9-117.9
Preoperative Diagnosis
No.
%
Primary osteoarthritis Osteonecrosis Posttraumatic arthritis Other
74 18 6 13
65 16 1 13
Results Fig. 1. Harris-Galante porous II acetabular components and stem.
Follow-up methods used in contacting patients were modeled after those used and well-described by King et al [3]. Families of deceased patients were contacted to find out if the patient had a revision before death. Patients returned for clinical follow-up at 3 months, at 1 year, and at 5-year intervals after surgery. Patients who had not undergone revision surgery returned for clinical follow-up 15 to 20 years after surgery. Follow-up included a physical examination with clinical and radiographic evaluation. Harris hip scores (HHSs) were collected preoperatively and at follow-up, with results classified as excellent (90-100), good (80-89), fair (7079), or poor (b70). Patients who were unable to return for a physical examination were administered the function portion of the HHS over the telephone. Radiographs of the operative hip taken at 3 months and at 15 to 20 years postoperatively in the anteroposterior and lateral views and were evaluated for evidence of implant failure. Specific radiographic evaluation included stem osteolysis, pedestal formation, implant-bone interface evaluation, and polyethylene wear. All radiographs were digitally scanned (Epson Expression 1680, Epson America, Inc, Long Beach, Calif). Radiographic analyses consisting of cup inclination angle, cup anteversion angle, linear penetration rate, and volumetric wear rate were measured using a commercially available software program (Polyware Pro 3D, Draftware Developers Inc, Vevay, Ind). All statistical analyses were conducted using SPSS (Version 13.0 for Windows, SPSS Inc, Chicago, Ill). Radiographic analyses were correlated using the Pearson product-moment correlation. A P value of less than .05 was considered significant. Implant survivorship was characterized using Kaplan-Meier analysis with revision for any reason and for specific reasons as end points. Implant success was defined as the prosthesis remaining in situ.
One hundred one patients (51 males, 50 females, 113 hips) comprised the index group. The distributions of demographic characteristics and preoperative diagnoses are displayed in Table 1. Of the 113 hips, the HGP I cup was used in 85 and the HGP II cup in 28. At follow-up, 32 patients (34 hips) were found to be deceased; 6 patients (9 hips) were lost to follow-up; 10 living patients (10 hips) had undergone revision surgery; and 53 patients (60 unrevised hips) returned for clinical evaluation at a mean of 17.2 years (range, 15-20 years). Of the 32 deceased patients, 11 of 34 hips were known to be unrevised at time of death, but revision status on the remaining 23 hips was unable to be ascertained. Revision arthroplasty on the 10 patients included 8 patients (9.5%) who received HGP I cups and 2 patients (6.9%) who received HGP II cups. The length of survival and reasons for revision surgery are displayed in Fig. 2. The average time to revision surgery was 10.74 years (range, 2.9217.75 years). The HHSs are displayed in Table 2. The function category of HHS is displayed separately to include patients who were not able to complete the entire scoring system. Radiographically, 12 hips demonstrated eccentric wear; 6 had femoral osteolysis (Fig. 3); 2 had acetabular osteolysis; and 1 had a broken tine (Fig. 4). A subgroup assessment using 2-dimensional analysis performed on 25 of the hips for which both baseline (3 months
Fig. 2. Time course and reasons for revision surgery during 15to 20-year follow-up.
HGP Femoral and HGP I and II Acetabular Components Anseth et al
689
Table 2. Harris Hip Scores and Clinical Outcome of Unrevised THA Patients at 15 to 20 Years No. HHS (0–100) Preoperative 49 At follow-up 39 Harris hip function score (0-47) At follow-up 48
Mean ± SD
Range
57 ± 15.5 82 ± 20.0
23-95 30-100
37 ± 9.5
11-47
Clinical outcome
No.
%
Excellent/Good Fair/Poor
25 14
64 36
The Harris hip function score (HHFS) is a subset of the HHS that can be administered by telephone. The 48 patients completing the follow-up HHFS include the 39 who were able to return for follow-up and completed the entire HHS plus 9 patients who could only complete the HHFS section by telephone.
postoperative) and 15- to 20-year follow-up radiographs were available demonstrated a mean linear polyethylene penetration rate of 0.153 mm/y and a mean volumetric
Fig. 4. X-ray depicting a broken tine and asymmetric poly wear in the HGP acetabular component.
wear rate of 74.96 mm3 (Table 3). More hips were not evaluated for analysis because of the lack of precision and quality of x-rays, which are necessary to do this measurement. A small difference in polyethylene wear rates between the 2 cups was present with 0.161 mm/y in the HGP I cup and 0.139 mm/y in the HPG II cup. No correlation was found between linear penetration rates and cup anteversion angle or cup abduction angle. Overall, survival status was established for 73 patients representing 81 of the 113 index hips. The survival rate (percentage of hips not requiring a revision) in these patients was 87.7% (71 of 81 were unrevised) at 20 years. Worst-case scenario calculation, with patients either lost to follow-up or of unknown revision status at time of death assumed to be revised, yielded a survival rate of 62.8% (71 of 113 unrevised). Best-case scenario calculation, with patients either lost to follow-up or of unknown revision status at time of death assumed to be unrevised, yielded a survival rate of 91.2% (103 of 113 unrevised). Using the Kaplan-Meier method to calculate implant survivorship, the probability of retaining the implant at
Table 3. Results of Radiographic Analyses Showing Linear Penetration Rate, Volumetric Wear Rate, and Cup Angle at 15 to 20 years
Fig. 3. X-ray demonstrating osteolysis around the femoral stem.
Radiographic Analysis
No.
Linear penetration rate Volumetric wear rate Cup abduction angle Cup anteversion angle
25 25 25 25
Mean ± SD 0.153 ± 74.96 ± 35.54° ± 20.26° ±
0.124 mm/y 75.01 mm3 9.54° 9.31°
690 The Journal of Arthroplasty Vol. 25 No. 5 August 2010
Fig. 5. Survivorship curves showing the probability of implant retention with all-cause and cause-specific revision as the end points.
20 year follow-up with revision for any reason as the end point was 84.7% (95% confidence interval [CI], 74.7%91.5%). The survivorship probabilities for acetabular aseptic loosening (n = 2), polyethylene wear (n = 2), and femoral aseptic loosening (n = 6) as end points were 97.4% (95% CI, 90.4%-99.5%), 96.7% (95% CI, 89.3%99.2%), and 90% (95% CI, 80.8%-95.2%), respectively (Fig. 5).
Discussion The longevity of the HGP acetabular components has been excellent in short- and long-term follow-up studies with regard to resistance to mechanical loosening. Our study documented outstanding survival of the HGP cup with only 2 failures due to acetabular aseptic loosening. Survivorship probabilities were 97.4% for acetabular aseptic loosening revisions and 96.7% for polyethylene wear revisions as end points, similar to what has been reported. One study documented no revision of acetabular components at 15 years because of aseptic loosening [4]. Another study documented 1 revision for aseptic loosening at 6 years [5], and 3 other studies documented no revisions of HGP acetabular components at 10 years for aseptic loosening, but 2 for osteolysis around the cup [2,6,7]. The HGP acetabular component's original weakness was the locking mechanism for securing the polyethylene liner. Several authors reported disassociation of liners from securely fixed acetabular cups [8-10]. Even in the long-term follow-up studies, polyethylene disassociation is a more common reason for revision of the acetabular component than aseptic loosening. Clohisy and Harris [7] reported in 2 different studies that acetabular revision due to liner dislodgement and tine fracture was the reason for 3 of 4 revisions in 1 report and 3 of 8 revisions in the other [6]. Parvizi et al [4] documented in their study that 2 HGP acetabular components were revised at 2 to 3 years because of acetabular liner disassociation. In 1 case, the cup was retained, but in the other, the cup also had to be revised. Tompkins et al [11] reported 3 revisions in a series of HGP hips, 2 for polyethylene exchange and 1
for “massive” retroacetabular osteolysis. In our study we did not find any clinically apparent liner disassociation and had only one radiographically visible broken tine. Several investigations have shown polyethylene wear rates from 0.015 to 0.18 mm/y with the HGP cups implanted during the same period [2,5,11-17]. Some investigations have found that as the wear rate increases, the HGP stem and cups become more susceptible to osteolysis and aseptic failure. D'Lima et al [5], following a cohort with a subset of patients also included in this study, were able to demonstrate a link between polyethylene wear and femoral osteolysis; a positive correlation between the 2 was detectable at a mean follow-up of 6 years, with polyethylene wear at 0.11 mm/y in hips with no femoral osteolysis and at 0.18 mm/y in hips with femoral osteolysis. Two other studies [2,16] documented greater osteolysis surrounding the HGP cups in which the polyethylene wear rate was 0.16 mm/y. Our study demonstrates good long-term survival with a mean polyethylene linear penetration rate of 0.153 mm/y. The difference in mean polyethylene linear penetration between HGP I and HGP II cups was 0.022 mm/y. The fact that a small number of x-rays were evaluated in this manner leaves this study open to a type II statistical error and, therefore, we did not present statistical differences. The other shortcoming of the early-generation HGP cementless THA was the femoral stem, a titanium alloy stem with commercially pure titanium fiber-mesh porous-coating ingrowth pads located on the anterior, the posterior, and the medial surfaces of the proximal third of the stem. These pads provide points of attachment for bony ingrowth, and the smooth channels between the pads provide access for wear particles to affect the femoral shaft. In our study, 6 hips were revised for femoral aseptic loosening before the follow-up period, and 2 hips had radiographically loose femoral components at the time of follow-up. The durability of the HGP femoral component was a concern even in early studies. One study [18] reported femoral lysis was present in 31% of the femurs at 53 months, and another study [15] found osteolysis in 21% of femurs at a mean follow-up of 6 years. Two studies [17,19] reported inferior survivorship of the HGP femoral component with an 86% survivorship at 10 years and a 76.3% survivorship at 13 years, respectively. Parvizi et al [4] reported 8 femoral revisions for aseptic loosening at a mean of 8.2 years and predicted femoral survival would be 82% at 15 years with mechanical failure as the end point, which is comparable to the 84.7% survivorship in our study. The Kaplan-Meier survivorship probability with femoral revision as the end point was 90% (95% CI, 80.8%-95.2%) in our cohort. Petersilge et al [13] in their prospective study, which includes a subset of patients also in this study, documented a postoperative mean HHS of 92 at 5.8 years in a population of cementless HGP hip arthroplasties. Archibeck et al [2] also found a high postoperative
HGP Femoral and HGP I and II Acetabular Components Anseth et al
HHS with second-generation cementless hips, reporting a mean of 94 at 10 years. Kim [20] and Ritter et al [21] documented a steady decrease in HHS over the postoperative period in cementless hips. When considering the known deterioration in HHS with time, our average HHS of 82 at a mean follow-up of 17.2 years is consistent with other long-term studies. As with many long-term longitudinal studies out to 20 years, a number of the patients were lost to follow-up. This was due either to patient death, patients' inability to return for health reasons, or an inability to locate patients. Locating and contacting patients this many years following the index surgery are challenges attested to by the literature [18]. The subgroup that was used to determine penetration wear rate was a small percentage of the entire study group and could have introduced a type 2 error. In our study, the worst-case survival of the HGP hip at 20 years with revision for any reason as the end point was 62.8%, which compares favorably to the 25-year results documented as 63% survival in the living Iowa cemented THA population [22]. Given some of the now apparent design flaws in this early generation of cementless prostheses, with regard to the acetabular locking mechanism and the femoral ingrowth surface, we can reasonably expect the next generation of cementless prostheses to surpass these survivorship numbers when they reach the same period of follow-up.
8.
9.
10.
11.
12.
13.
14.
15.
16.
References 1. Clohisy JC, Harris WH. The Harris-Galante porous-coated acetabular component with screw fixation. An average tenyear follow-up study. J Bone Joint Surg Am 1999;81:66. 2. Archibeck MJ, Berger RA, Jacobs JJ, et al. Secondgeneration cementless total hip arthroplasty. Eight to eleven-year results. J Bone Joint Surg Am 2001;83-A:1666. 3. Rothman RH, Cohn JC. Cemented versus cementless total hip arthroplasty. A critical review. Clin Orthop Relat Res 1990;254:153. 4. Parvizi J, Sullivan T, Duffy G, et al. Fifteen-year clinical survivorship of Harris-Galante total hip arthroplasty. J Arthroplasty 2004;19:672. 5. D'Lima DD, Yashar AA, Venn-Watson EJ, et al. The HarrisGalante porous acetabular component at intermediate follow-up. Orthopedics 2001;24:747. 6. Clohisy JC, Harris WH. The Harris-Galante uncemented femoral component in primary total hip replacement at 10 years. J Arthroplasty 1999;14:915. 7. Clohisy JC, Harris WH. Primary hybrid total hip replacement, performed with insertion of the acetabular component without cement and a precoat femoral component
17.
18.
19.
20.
21.
22.
691
with cement. An average ten-year follow-up study. J Bone Joint Surg Am 1999;81:247. Diekerhof CH, Van Der Hart CP, Fidler MF. Dislocation of polyethylene liner in uncemented Harris Galante II acetabulum: report of 6 cases. J Arthroplasty 2004;19:661. Star MJ, Colwell Jr CW, Donaldson III WF, et al. Dissociation of modular hip arthroplasty components after dislocation. A report of three cases at differing dissociation levels. Clin Orthop Relat Res 1992;111. Gonzalez dV, Ruzo PS, Li S, et al. Dislodgment of polyethylene liners in first and second-generation HarrisGalante acetabular components. A report of eighteen cases. J Bone Joint Surg Am 2001;83-A:553. Tompkins GS, Jacobs JJ, Kull LR, et al. Primary total hip arthroplasty with a porous-coated acetabular component. Seven-to-ten-year results. J Bone Joint Surg Am 1997;79: 169. Cannestra VP, Berger RA, Quigley LR, et al. Hybrid total hip arthroplasty with a precoated offset stem. Four to nine-year results. J Bone Joint Surg Am 2000;82:1291. Petersilge WJ, D'Lima DD, Walker RH, et al. Prospective study of 100 consecutive Harris-Galante porous total hip arthroplasties. 4- to 8-year follow-up study. J Arthroplasty 1997;12:185. Brown III EC, Lachiewicz PF. Precoated femoral component in total hip arthroplasty. Results of 5- to 9-year followup. Clin Orthop Relat Res 1999;153. Latimer HA, Lachiewicz PF. Porous-coated acetabular components with screw fixation. Five to ten-year results. J Bone Joint Surg Am 1996;78:975. Soto MO, Rodriguez JA, Ranawat CS. Clinical and radiographic evaluation of the Harris-Galante cup: incidence of wear and osteolysis at 7 to 9 years follow-up. J Arthroplasty 2000;15:139. Cruz-Pardos A, Garcia-Cimbrelo E. The Harris-Galante total hip arthroplasty: a minimum 8-year follow-up study. J Arthroplasty 2001;16:586. Smith E, Harris WH. Increasing prevalence of femoral lysis in cementless total hip arthroplasty. J Arthroplasty 1995; 10:407. Thanner J, Karrholm J, Malchau H, et al. Poor outcome of the PCA and Harris-Galante hip prostheses. Randomized study of 171 arthroplasties with 9-year follow-up. Acta Orthop Scand 1999;70:155. Kim YH. Long-term results of the cementless porous-coated anatomic total hip prosthesis. J Bone Joint Surg Br 2005;87: 623. Ritter MA, Thong AE, Davis KE, et al. Long-term deterioration of joint evaluation scores. J Bone Joint Surg Br 2004;86:438. Keener JD, Callaghan JJ, Goetz DD, et al. Twenty-five-year results after Charnley total hip arthroplasty in patients less than fifty years old: a concise follow-up of a previous report. J Bone Joint Surg Am 2003;85-A:1066.