Minimum 10-year Results of a Tapered, Titanium, Hydroxyapatite-Coated Hip Stem: An Independent Review

Methods 

Two visiting surgeons who were not involved in the initial or subsequent care of the patients independently evaluated the experience of a single surgeon using a consecutive series with 1 type of femoral implant at 1 orthopedic center. A careful review of a prospective database and the personal implant logs of the surgeon was conducted, and results are presented for the first 147 consecutive hips in 133 patients who received a hip arthroplasty using a tapered, titanium (Ti6Al4V), HA-coated femoral implant (Corail; Depuy Orthopaedics Inc, Warsaw, Ind) (Fig. 1). This experience includes all patients who received this implant between August 1986 and December 1988 at this center.

View Large Image Download to PowerPoint

Fig. 1. Corail femoral implant: note full HA coating, horizontal and vertical normalization, and tapered geometry.

This femoral implant is available with a collared option, which was used in 33 hips. The remaining 114 hips in this series were inserted without a collar. The titanium tapered implants were entirely coated with a layer of 155 ± 35 μm of HA, applied with a plasma spray technique. The sizes of the implants used in this series were as follows: 26 size 10, 30 size 11, 32 size 12, 30 size 13, 19 size 14, 8 size 15, 1 size 16, and 1 size 18. Femoral heads used included 32 mm in 112 hips and 28 mm in 35 hips. Although the femoral implant was consistent during this experience, a variety of acetabular implants were used, and therefore, their performance was not the target of this investigation.

The average age was 63 years (range, 41-87 years) among the 70 men (78 hips) and 63 women (69 hips). The mean height was 162 cm (range, 154-195 cm) and the mean weight, 71 kg (range, 54-95 kg). The diagnosis was osteoarthritis in 106 hips, osteonecrosis in 14 hips, dysplasia in 10 hips, inflammatory arthritis in 7 hips, and other conditions in 11 hips. Sixteen hips had undergone prior procedures, including osteotomy in 6, osteosynthesis of femoral neck fracture in 5, acetabuloplasty in 2, and unspecified in 3. Patients were classified by activity level and Charnley class [13], with 42 hips replaced in heavy laborers, 39 in moderate laborers, 32 in light laborers, 31 in semi sedentary, and 3 in sedentary individuals. There were 95 hips done in patients classified as Charnley class A (1 hip was involved, but no other condition interfered with walking), 46 class B (both hips were involved, but body was normal and does not affect walking), and 6 class C (additional factor contributing to failure of normal walking ability).

Each surgery was performed through an anterolateral approach to the hip with the patient supine. Implants were available in 1-mm increments from 9 to 18 mm and were implanted using compaction broaching. Using this technique, the anteversion was established and the cancellous bone compacted toward the cortical rim until axial and rotational stability was achieved. Cancellous bone was preserved and no attempt was made to have the implant contact the femoral cortex. Rather, the HA-coated femoral stem was implanted into a compacted sleeve of cancellous bone.

Patients were allowed to bear weight as tolerated postoperatively and were protected with crutches as necessary. All patients had routine perioperative antibiotics and thromboembolism prophylaxis with subcutaneous heparin.

All patients were prospectively entered into a database, and clinical and radiographic data were collected postoperatively. Although a regular regimen of postoperative follow-up was planned, patient compliance was poor. Preoperative and early postoperative data were available for all patients, but long-term data were missing from the database. Therefore, at the initiation of this investigation, the database was scrutinized for missing data, and patients for whom 10-year data were lacking were identified. These patients were contacted and asked to return for a clinical and radiographic evaluation. Those who would not return were asked to obtain radiographs and send them for review.

Clinical evaluation was performed using the scoring system of d'Aubigne and Postel [14], and hip scores were assigned according to level of pain, functional status, and range of motion. Patients who refused to return, but who did forward x-rays for review after being contacted, were questioned by phone about the functional status of their hip.

All available radiographs were collected and assessed for implant stability, subsidence, osseointegration, osteolysis, stress shielding, and evidence of periprosthetic lucency. Anteroposterior and lateral views of each hip were obtained preoperatively, postoperatively, and at routine follow-up. Early postoperative views were compared with films taken at final follow-up to assess subsidence, acetabular wear, and bone remodeling. Subsidence was assessed by measuring any change in the distance from the tip of the greater trochanter to the lateral shoulder of the femoral stem on sequential radiographs. Varus valgus positioning and subsidence were determined by comparing the mid axis of the femoral shaft vs mid axis of the implant on sequential films. A zonal analysis, as outlined by Gruen et al [15] and Johnston et al [16], was used to catalogue relevant changes in bone morphology and bone implant interface characteristics. Osseointegration was determined by the presence of spot welds.

Results 

Of the 133 patients (147 hips) in this study, 40 patients (44 hips) were deceased at the time of this review without 10-year follow-up having been obtained. Intermediate clinical and radiographic follow-up ranging from 2 to 7 years was available for 35 of these deceased patients, accounting for 38 hips. These data were analyzed in the manner outlined hereinabove, and all stems in this group were well fixed and exhibited good evidence of stability. There were no impending failures. The remaining 5 patients with 6 hips had died with their hips intact, but no clinical or radiographic data were available, beyond 1-year follow-up. There were no known revisions in the group of deceased patients and no known functional or clinical hip problems. Another 7 hips in 7 patients were considered lost to follow-up because no information could be obtained about their status, beyond their initial postoperative check, at which time, the implant was functioning well.

As a result, 96 hips (86 patients) were available for review at follow-up of greater than 10 years. Of these, 75 hips were evaluated both clinically and radiographically at an average of 11.5 years (range, 10-13.5 years) follow-up. In 15 of the 96 cases, acetabular failure required revision of the acetabular component, but the femoral stem survived and was available for long-term evaluation. In 2 additional cases, both acetabular and femoral revision was performed. In both cases of femoral revision, the femoral component was well fixed, with considerable osseointegration, and a femoral osteotomy was required for removal. Radiographs were obtained at 10-year follow-up for another 19 hips, but the patient would not come in for the 10-year clinical evaluation. Therefore, a phone interview was conducted to assess any change in functional status at a minimum of 10 years. Patients were questioned in detail about possible hip-related symptoms and change in functional status. None of these patients admitted to a problem with their hip and refused to return for clinical and radiographic documentation because they lived far away and were not having a problem. None of these patients had had, or expected, further surgery on their hip.

None of the 96 implants that were directly accounted for were revised for aseptic loosening of the femoral component, and there was no evidence of aseptic loosening in any follow-up radiograph. There was no distal osteolysis, although 19 (19.8%) had osteolysis confined to zones 1, 7, and 8 (Fig. 2), including all 15 cases that had undergone acetabular revision for lysis. In all cases, the osteolysis was continuous with the joint space and was limited to the proximal zones. It became encapsulated by a sclerotic margin and was slow to enlarge. Osteolysis was never a structural problem in the femoral periprosthetic bone. The acetabular revisions were due to polyethylene wear and osteolysis. In these cases, there was acetabular loosening and severe acetabular osteolysis. Yet, despite this heavy burden of polyethylene particulate debris, osteolysis was never a clinical or radiographic problem on the femoral side.

View Large Image Download to PowerPoint

Fig. 2. Osteolysis: when present, confined to proximal zones 1 and 7.

Bone remodeling around the implant was generally favorable. There were 5 hips with incomplete radiolucencies, none affecting more than 30% of the interface. Calcar remodeling was common, with rounding of the calcar in 50% of hips. The group with a collared prosthesis exhibited significantly less calcar rounding. Interestingly, calcar remodeling with bone hypertrophy was present in cases with incomplete seating of the implant. When HA coating was left exposed above the calcar, it was common to see creeping deposition of new bone along the HA coating, yielding an increase in calcar height, suggesting a significant osteoconductive quality.

All femoral implants had evidence that the HA coating had stimulated an osteogenic response, exhibiting evidence of osseointegration, with the so-called spot welds or periprosthetic diaphyseal endosteal bone formation seen in at least 3 zones (Fig. 3); zones 2 and 3 were the most reliable sites of spot welds. Subsidence was seen in 5 hips and was less than 4 mm in each case. Stress shielding was significant in only 2 cases in which the implants were preferentially bonded distally. The remainder exhibited normal or near-normal preservation of proximal bone structure (Fig. 4).

View Large Image Download to PowerPoint

Fig. 3. Periprosthetic osteogenesis develops reliably around implant.

View Large Image Download to PowerPoint

Fig. 4. Long-term (>10 years) follow-up demonstrates bone preservation.

Clinical scores indicated that postoperative hip function improved dramatically when compared with preoperative levels. Preoperative composite scores were 5.5 of 18, whereas at latest follow-up, scores averaged 16.5 of 18. Scores improved in the first year and were maintained throughout the study period, unless acetabular failure intervened. Patients requiring acetabular revision had lower scores at final follow-up, but this was not statistically significant.

There was no complaint of thigh pain in any patient in this series. A limp was present in 12 patients evaluated at greater than 10 years, and 8 patients were using an ambulatory aide, 6 of whom were semisedentary preoperatively.

Discussion 

This is a study of a prospectively collected series of hip arthroplasty using a tapered, titanium, HA-coated hip stem; the results of which were analyzed independently by surgeons who did not perform the original surgery. There was a high rate of follow-up of 96 hips (86 patients) with only 7 patients from the original cohort that could not be accounted for. The femoral implant evaluated in this study has been shown to perform quite well historically [17], and in this group of patients, it developed osseointegration with clear evidence in all cases of endosteal spot welds and trabecular bridging. There were no cases of aseptic loosening, with a stable bone response demonstrated reliably at greater than 10 years of follow-up. There was no evidence of subsidence or component migration, indicating that implant stability was achieved early and reliably.

The results of this study are comparable to those reported by Røkkum and Reigstad [18] using this same device. In their review of 94 consecutive cases, with careful radiographic and clinical follow-up, there was no observed stem subsidence or loosening in any case. The authors regularly found cancellous bone remodeling in the proximal femur, with preservation of the osseous architecture. Reikeras and Gunderson [19] reported on the performance of this stem at 8 to 12 years of follow-up, finding only 1 case of mechanical failure of 245 inserted stems. Proximal bone atrophy and distal hypertrophy were infrequent, leading the authors to postulate an “essentially physiologic” load profile. Proximal osteolysis was minimal and distal osteolysis absent, despite a high rate of acetabular failure, similar to that seen in our cohort. In a retrieval analysis of this implant, Tonino et al [20] found that all 5 stems were well fixed in the femur, and there was a significant evidence of both distal and proximal osseointegration.

Although the relative contribution of the hydroxyapatite coating to the success of this implant cannot be determined by this study, there is considerable clinical and experimental evidence suggesting that such coatings enhance initial fixation and ultimate osseous integration 8, 11, 21. Comparative studies have indicated advantages of HA-coated implants over noncoated implants of similar design 9, 22, 23, 24, 25, with HA-coated femoral stems achieving more reliable bony fixation, better evidence of spot welds, and less subsidence than porous control groups. Using dual-energy x-ray absorptiometry, Tanzer et al [25] found that after 2 years of follow-up, stems with HA coating exhibited significantly less bone loss than stems implanted without HA coating. In addition, there are a number of reports documenting the long-term success of HA-coated femoral components at greater than 10 years of follow-up 5, 26, 27, 28, 29, with absences of fixation failure of the femoral stem seen in 96% to 100% of cases.

The tapered geometry of this implant may also have played a significant role in the osseous integration and bone adaptation that was observed in this study. Long-term success has been documented with other tapered stems, without hydroxyapatite coating 12, 30. Although it has been postulated that tapered femoral components optimize the stress transfer from implant to bone, this study cannot substantiate that claim. Comparison of early postoperative with long-term radiographs, however, does demonstrate preservation of bone architecture, with similar appearance of bone quality, both early and late. In addition, the complete absence of thigh pain noted in this group of patients may be related to this favorable bone response without significant hypertrophy or remodeling.

One particularly relevant aspect of the current study is the excellent femoral performance despite the high number of acetabular failures (16%) in this series and the high burden of polyethylene debris. Most acetabular failures were secondary to polyethylene failure with granuloma formation and secondary loosening. The Atoll cup used in most of these cases had a thin polyethylene insert that was gamma sterilized in air. Poor results with this design have been previously documented 31, 32, with Lai et al [31] similarly finding a survival rate of 97.5% for HA-coated femoral stems, despite an acetabular cup failure rate of 32.1%. Although it is possible to invoke the presence of HA particles within the interface as a potential source of increased wear [33], the present study cannot substantiate that possibility. It is noteworthy that despite the high rate of acetabular wear, there were no cases of significant femoral osteolysis. There was no distal osteolysis; when present, lysis was confined by a sclerotic zone to the proximal zones of the implant and was not progressive. It is likely that the circumferential and complete coating of the implant conferred upon it a resistance to progressive and distal manifestations of osteolysis.

In summary, this study documents the success of this tapered, titanium, HA-coated femoral stem in a consecutive series of unselected cases with no failure of fixation or aseptic loosening. The implant was found to perform with equal success across a wide range of pathologic entities, patient profiles, and bone types. Although the relative contributions of the surface coating, the implant geometry, and the surgical technique of compaction broaching cannot be established by this study, it is clear that this combination of features resulted in a durable and reliable femoral construct for this group of patients. Further documentation of the performance of this device into the second decade of service and beyond will be necessary to determine if these favorable results are maintained.