The long-term outcome of the cemented Weber acetabular component in total hip replacement using a second-generation cementing technique

The debate on cemented versus uncemented acetabular components in total hip replacement (THR) continues. Although some authors suggest abandoning the use of cemented implants,¹ others show good survival rates of 90% after 25 years.² The survival rates in these studies vary, but the common conclusion is that aseptic loosening is the main cause for revision in cemented acetabular components. The Scandinavian arthroplasty registers^3–6 show that in young patients, there is no significant difference in the overall survival rate between uncemented and cemented acetabular components with revision as the endpoint. For all-polyethylene acetabular components the ten-year survival was 87% (95% CI 85 to 90), and for porous-coated uncemented components 88% (95% CI 85 to 90), with aseptic loosening as the main reason for revision when cemented components were used. Uncemented components were also exchanged for other reasons, such as polyethylene wear and osteolysis (Finnish,³ Norwegian,⁴ Danish⁵ and Swedish⁶ Hip Registers).

Aseptic loosening is influenced by the cementing technique used,⁷ and it seems clear that this needs to be optimised. Several reports have been published on improvements in the technique of cementing of the femoral component,^8–12 but only a few papers have addressed the acetabular side.^13–17

We modified the cementing technique for the acetabular component in 1978.¹⁸ This was first presented at a symposium in 1982 and published in 1982.¹³ It was anticipated by the senior author (RKM) that the technique would improve the survival of the implant, and this has been demonstrated by comparing survival rates of patients before and after 1978 in an earlier study.¹⁸ We now again describe this method and the long-term results. We have also compared our results to the long-term outcomes of other cemented acetabular components reported in the literature.

Patients and Methods

Between 1978 and 1993, the senior author (RKM) performed 287 THRs in 244 patients, 65 of whom were men, using a cemented hemispherical Weber acetabular component (Allopro, Baar, Switzerland) implanted with a technique developed by him. In order to obtain a clear view of the outcome using this technique the number of variables were reduced by excluding operations performed by orthopaedic trainees and those where any other acetabular component was used.

The mean age of the patients at the time of surgery was 65.3 years (21 to 90). The reason for the THR was idiopathic osteoarthritis in 140 hips, osteoarthritis secondary to dysplasia in 96, rheumatoid arthritis in eight, avascular necrosis of the femoral head in 23, post-traumatic osteoarthritis in 13, and other secondary osteoarthritis in seven.

Previous surgery had been performed in 73 hips and included an intertrochanteric osteotomy in 56 patients, a pelvic osteotomy in two, a shelf-plasty in two, and an internal fixation of a proximal femoral or acetabular fracture in 13.

Operative technique and implants.

All procedures were done with antibiotic prophylaxis, using an anterolateral approach (Watson-Jones)¹⁹ with the patient in the supine position. The Weber Rotation (Allopro) THR was used as a standard implant in all primary cases. The femoral component was manufactured from Protasul 10 alloy (Sulzer AG, Wintherthur, Switzerland) with a cylindrical trunnion made of Protasul 2 (Sulzer AG) which permits rotation. To this trunnion 32 mm diameter heads were attached, of which 100 were made of Protasul and 187 of ceramic (Biolox, Feldmühle, Plochingen, Germany).

The head was placed on the rotating Protasul 2 cylinder, which was available in three lengths (42 mm, 47 mm and 52 mm). The Weber hemispherical acetabular component was used in every patient. This all-polyethylene component (RCH-1000 Chirulen, Hoechst, Hostalen, Germany) was available with an external diameter ranging between 40 mm and 64 mm, and a depth which varied between 24 mm and 37 mm. The rim of the component had a scalloped profile. The advantage of this profile was to allow partial or complete removal of the prominences giving a perfect fit of the component in the reamed acetabulum, creating intrinsic stability before cementing (Fig. 1).

The interrupted shape also allowed compression of the cement between the prominences using a special impaction device during polymerisation of the cement. A 47 mm diameter component was used in 62 hips (21.6%), the 52 mm in 179 hips (62.4%), the 57 mm in 44 (15.3%) and the 64 mm in two (0.7%).

Cementing technique.

Our technique for cementing the acetabular component involves reaming the acetabulum to the depth of the subchondral layer, making six to eight multiple anchorage holes, which are conical in shape and have a diameter of 8 mm. Only the hard superficial bone of the subchondral plate is drilled. The cancellous bone beneath the anchorage holes is impacted instead of being removed, using a specially designed round-ended impactor with a diameter of 8 mm. Anchorage holes are produced to a depth of 0.5 cm to 1.5 cm, depending on the softness of the bone (Fig. 2). This method lines the holes with a homogenous layer of cancellous bone.

When the quality of the cancellous bone is poor and subchondral cysts are present, cancellous bone from the femoral head is impacted into the anchorage holes. In the presence of a very sclerotic acetabulum, a 6 mm diameter drill is used to roughen the bony surface in between the holes.

For implantation of the acetabular component the anchorage holes are dried and filled with multiple gauze swabs. The swabs are removed, quickly followed by digital introduction of low-viscosity Sulfix cement (Sulzer AG) into the holes, to avoid any bleeding between the cancellous bone and the cement, creating optimal compression of the cement-bone interface. The cement is not handled until initial polymerisation has commenced and it is losing its stickiness. This can be determined by the cement losing its shiny surface, which generally occurs after six to eight minutes. The cement must reach this phase to permit proper compression. The polyethylene acetabular component is then introduced and the cement is primarily compressed in the direction of the craniolateral anchorage holes. Shortly afterwards, anteversion and inclination are corrected, still under compression. By pressing the component in the craniolateral direction first, extrusion of cement on the medial side can usually be avoided and high pressurisation of the cement into the important anchorage holes is achieved.

Approximately 40% of the full 40 g mix of cement is usually required to achieve good fixation.

A trochanteric osteotomy was performed in 48 hips. In 91 hips, bony cover of the acetabular component was insufficient, requiring the addition of an acetabular shelf-plasty. Bone graft was added to improve the bony cover of the acetabular component in cases where the acetabulum was dysplastic and the craniolateral rim was lacking.²⁰

Follow-up.

The patients were seen at regular intervals, with an initial post-operative visit after six weeks, and then at three, six and 12 months, two years, and biennially thereafter. The follow-up consisted of radiological and clinical evaluation. The Harris hip score (HSS)²¹ was used to assess the clinical outcome. For the radiological analysis, weight-bearing anteroposterior (AP) pelvic and lateral radiographs were obtained and were scrutinised for radiolucent lines around the component and signs of migration, or change of inclination in the case of the acetabular component, using the criteria described by Harris et al¹¹ for the femoral and that by Hodgkinson, Shelley and Wroblewski²² for the acetabular component.

A descriptive analysis was performed for the complications, clinical and radiological follow-ups. Survival analysis was calculated using the life-table method. The primary endpoints were revision for radiologically-proven aseptic loosening of the acetabular component and radiological evidence of loosening. Secondary endpoints included revision for any reason, revision for aseptic loosening of the femoral component, revision for loosening of either of the components, and definitive signs of radiological loosening of the femoral component.

In order to evaluate the success of this technique in younger patients, we divided the group of 278 hips into patients under 55 years of age (39 hips) and those over that age (239 hips) and analysed them separately.

Results

A total of 21 hips in 20 patients were revised. Apart from the acetabular revisions, five patients needed a revision of the femoral component where the acetabular component was left in place. In addition one patient had an isolated revision of the stem after 9.1 years with a further revision involving the acetabular component ten years later (19.1 years after the primary procedure). Of the six femoral component revisions performed, one stem was revised because of a peri-prosthetic femoral fracture 14.1 years after implantation.

Aseptic loosening necessitated isolated revision of the acetabular component in four patients, and eight patients had a revision for aseptic loosening during which both components were exchanged. Two patients had septic loosening of their components, resulting in the removal of both.

At the latest follow-up, 107 of the 224 unrevised patients (117 hips) had died at a mean of 9.6 years after surgery (0.01 to 25). Two patients (two hips) died shortly after surgery due to a myocardial infarction. In total, 35 patients (44 hips) were lost to follow-up, but were included in the survival and radiological analyses until their last review at a mean of 7.9 years (0.1 to 18.6) after surgery.

With revision for aseptic loosening of the acetabular component as the endpoint, the ten-year survival was 99.1% (95% CI 97.9 to 100), the 15-year survival rate was 95.7% (95% CI 92.2 to 99.3) and the 20-year survival rate was 85.5% (95% CI 74.7 to 96.2) (Table I, Fig. 3).

In the 34 patients under 55 years of age at implantation (39 hips), survival of the acetabular component with revision for aseptic loosening as the endpoint was 97.1% (95% CI 91.4 to 100) at ten years, dropping to 81.7% (95% CI 64.6 to 98.7) at 16 years, when 14 hips were at risk. Radiological analysis showed 12 acetabular components with signs of possible loosening after a mean of 15.4 years (8.4 to 22.9), but none showed signs of definitive loosening (Table II; Fig. 3).

Radiographs of the femoral components showed six cases with possible signs of loosening after a mean of 16 years (13.2 to 19.3), one with probable signs after 22.9 years, and five with definitive signs of loosening after a mean 16.5 years (9.2 to 20.4). The survival for all the hips at various endpoints is presented in Table III.

In all, four patients needed a second operation without exchange of either component, one for heterotopic ossification causing ankylosis and three for nonunion after osteotomy of the greater trochanter.

The HHS was obtained from 93 patients (112 hips) at a mean of 14.9 years (10 to 23.7). The mean score was 91.5 (24 to 100). The HHS was under 70 in four hips, 70 to 80 in ten, 81 to 90 in 27, and 91 to 100 in 71.

The total number of per- or peri-operative complications was 22. Two femoral shafts were perforated during surgery. This was recognised but needed no additional action because the damage was only minor. In six THRs a crack occurred at the greater trochanter, in five of which there had been a previous intertrochanteric osteotomy. The greater trochanter was reattached with screws and cerclage wire. There were two sciatic nerve palsies which both made a full recovery. A haematoma developed in nine hips, five of which were treated with surgical debridement. Two wound infections healed with antibiotics and surgical debridement. One hip dislocated 14 years post-operatively, and remained stable after closed reduction.

Discussion

Our study shows that excellent long-term results can be obtained with cemented acetabular components when using our cementing technique.

A recent meta-analysis of studies comparing cemented and uncemented fixation in THR has shown that cemented fixation still outperforms uncemented fixation in large subsets of the study populations.²³ Although cementless fixation of acetabular components seems to reduce the rate of aseptic loosening, it has been reported that loosening frequently coincides with loss of acetabular bone.^24–26 Polyethylene wear and osteolysis have caused concern in uncemented acetabular components.^15,27–29 Apart from wear debris as a causative factor in peri-acetabular osteolysis, it has been proposed that stress shielding from a metal-backed device may contribute to bone resorption around the acetabular component.^30–32 Earlier studies showed a more uniform distribution of stresses to the peri-prosthetic bone with metal-backed components,³³ but more recent work describes an apparent mismatch between the elastic modulus of the metal backing and the peri-prosthetic bone, as well as a difference in structural stiffness of the implant.^34,35

A 3D finite element study by Manley, Ong and Kurtz³⁶ showed a more even distribution of stresses in the model when using polyethylene rather than chrome-cobalt metal-backed implants. The model suggested that with the use of polyethylene as an implant, peripheral bone resorption would occur, but with the potential for bone formation at the dome.

Our reported survival rate for the cemented Weber acetabular component using our cementing technique, and the latest results³⁷ of the Charnley acetabular component of 96.9% (95% CI 96.5 to 97.3) at eight years using more modern techniques, indicate that there is still room for the use of a cemented acetabular component.

An important step in our technique is the impaction of cancellous bone into the anchorage holes. This reduces bleeding from the bone and assists in the production of a dependable cement mantle.

We used finger-packing to insert the cement into the anchorage holes of the acetabulum. Only one randomised prospective study using Roentgen Stereophotogrammetic (RSA). Analysis compares finger-packing with a pressurisation technique.³⁵ It showed that the pressurisation method was superior to finger-packing. However, the method of finger-packing analysed was not completely comparable to our method, in which low-viscosity cement is used at a relatively late stage so that it performs like a high-viscosity cement, allowing digital pressurisation into each keying hole. Flivik et al³⁸ pressurised each individual anchorage hole with a special device before pressurising the cement in the reamed acetabulum. In their finger-packing group, the anchorage holes were not separately filled with cement, but the whole acetabulum was filled with a cement gun and then digitally compressed. The impacted bone in the anchorage holes and preservation of the subchondral layer prevents cement from leaking into the peri-acetabular cancellous bone. We believe this preserves the elasticity modulus of the bone around the implant, and when acetabular component loosening prevails, limited destruction of the bone stock occurs.

The role of anchorage holes was investigated by Mootanah et al³⁹ using a finite element analysis. They showed that the depth and size of the anchorage holes were of less importance than their inclination, and showed the optimal anchorage hole to be perpendicular to the acetabulum. This is the technique we employ.

We used both ceramic and metal heads, but have not analysed these separately. However, we do not believe this to have resulted in any difference in outcome. Although Schuller and Marti⁴⁰ showed a difference in the amount of wear between these two heads, it was established in an earlier study that this did not lead to a significant difference in the rate of loosening.¹⁸

Our study shows that excellent results can be obtained with hemispherical all-polyethylene cemented acetabular components. Unfortunately, the described Weber component with an interrupted rim and the Weber Rotation femoral component are no longer available. We anticipate similar results from the Weber fix system (Allopro), although it produces more wear than the Rotation system.⁴¹ We believe that our cementing technique offers dependable acetabular fixation in cemented THR with preservation of bone stock.

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.