7.4 Total hip replacement: implant fixation

Cemented fixation has been used successfully in total hip replacement for over 40 years, but uncemented implants were developed because of the higher rates of aseptic loosening noted in young, active patients. In 1987 Hungerford mistakenly attributed this mode of failure to the method of fixation and coined the term ‘cement disease’ but it was later realized that the cause was, in fact, osteolysis resulting from the generation of polyethylene wear debris.

There remains much debate over which type of fixation—cemented, uncemented, or hybrid—offers the most reliable and durable results. As technology and manufacturing processes have developed, the results for uncemented fixation have improved, but at the same time cementing techniques have advanced.

The basic arguments over fixation have been overtaken by tribological issues, but discrepancies remain with regard to the results published in the literature, usually of a single implant or technique from a single centre by the surgeon who designed it, and the findings of the national joint registries. Issues such as age, weight, activity, pathology, bone quality, canal morphology, experience, and cost, must always be considered and there are differences between the philosophies and performance of individual implants within the cemented and uncemented categories.

Wear, leading to the development of osteolysis, is the ultimate determinant of outcome in modern total hip arthroplasty, but neither implant can be studied in isolation, because it is the performance of the entire hip system (implants, bearing, and cement, if used) that will determine the function and longevity of the reconstructed hip.

Polymethylmethacrylate (PMMA) bone cement has been used in arthroplasty surgery for over 40 years, with many publications and registries confirming excellent long-term results, even in young patients. The constituents and properties of the successful cements have remained essentially unchanged, but the ways in which they are prepared and used have improved (Figure 7.4.1).

Cement consists of a powder and a liquid (Box 7.4.1), which, when mixed together, produce an exothermic reaction (40–46°C), with polymerization as the cement hardens. Modern cements contain radiopacifiers and may also contain chlorophyll to impart a green colour in order to differentiate cement from the bone.

The type of cement used is not simply an afterthought and the Swedish registry has confirmed that the specific cement used has a significant bearing on the risk of revision. Cements are available with a variety of setting times and viscosities, with the high viscosity cements having a lower reported revision rate. Most contain antibiotics, usually gentamicin, which has been shown to further reduce the risk of revision.

If revising an infected arthroplasty the cement can be customized, based on the sensitivity of the infecting organism, by adding appropriate heat-stable antibiotics in powder form (less than 10% of the weight of the powder), and such cement can also be used to fashion a temporary spacer or beads for use in a two-stage procedure.

Bone cement has no adhesive properties and works as a grout, rather than a glue. During its insertion and working phase the cement is pressurized and interdigitates with the cancellous bone, hardening and creating a micro-interlock, forming a cement–bone construct, giving immediate stability. Meticulous preparation of the bony bed is essential and cancellous bone must be retained.

Pulse lavage removes debris or loose bone and the canal is dried to improve interdigitation and shear strength. Intramedullary bleeding is reduced by the local application of hydrogen peroxide, epinephrine, or chilled saline and by hypotensive anaesthesia.

The optimum cement mantle should be even and 2–4mm thick around the entire component, without any gaps, which would weaken it and allow wear debris to access the bony interface. Two interfaces are created (cement–bone and cement–implant), with cement being strongest in compression (93MPa) but weaker in shear (42.2MPa) or tension (35.3MPa).

The quality of the cement is important and the fatigue strength is increased by vacuum mixing or centrifuging which decrease the porosity. During cement insertion inclusions, voids, or laminations of fat, blood, or air should be avoided as this can reduce its strength by a factor of five.

A cement restrictor is placed 2cm beyond stem tip and a hollow centralizer should be used. Pressurization is maintained throughout to negate the effects of back bleeding, which forces the cement back out of the cancellous bone, and to take into account the initial cement expansion during the exothermic reaction and subsequent overall shrinkage (3–5%).

Cement must be stored at a constant temperature and should not be preheated or chilled because this makes its setting time unpredictable. Theatre temperature is important because a 1°C rise can reduce the setting time by up to 1min.

The type of cement, theatre, and storage environment and the quality of the cementing technique (Table 7.4.1) are crucial to the longevity of the arthroplasty, with the methods of bone preparation, cement insertion, and implantation differing between acetabulum and femur. Cemented fixation is demanding and time and care must be taken to ensure that all of the stages are performed as well as possible on each and every occasion (Box 7.4.2).

The aim of uncemented fixation is to achieve initial press-fit stability with ‘macro-fixation’ of the component against the surface of the bone, which is prepared with a degree of under-reaming. Long-term stability is achieved by ‘micro-fixation’ with the subsequent ingrowth of the surrounding bone into the porous surface modifications of the implant. Hydroxyapatite (HA) can also be used to encourage this process, but initial stability is crucial, because excessive micromotion will prevent bony ingrowth.

Some early designs of femoral prosthesis were manufactured with gaps or surface irregularities to allow bone to grow through and anchor the prosthesis. One familiar example is the Austin Moore hemiarthroplasty, but in young or active patients this offered limited support and led to loosening. In others, HA was used to coat components, which had no other surface modifications, with the risk of the implants subsequently loosening if the coating separated from the implant or had been reabsorbed.

Modern uncemented implants feature coatings to part or all of their surfaces to encourage bony ingrowth, which will then provide long-term fixation. This process occurs within 6 weeks of implantation and some surgeons keep their patients partially weight bearing during this period. The initial fixation may be augmented by longitudinal flutes, spikes, fins, or screws (Figure 7.4.2).

The manufacturing of uncemented implants is more complex and time consuming than for cemented ones, as a result of which they are generally more expensive, but without significant improvements in prosthesis survival they will not be cost-effective.

Cemented high density polyethylene (HDP) acetabular components were popularized by Sir John Charnley as part of the low-frictional torque concept. The material of choice subsequently became ultra-high-molecular-weight polyethylene (UHMWP), with cementable highly cross-linked polyethylene (HXLP) implants only recently being made widely available, despite the fact that this material has been used in modular uncemented cups for many years.

Modern components incorporate a long posterior wall to reduce the risk of dislocation and a flange to improve cement pressurization, implant positioning, and to act as a potential barrier to wear debris (Figure 7.4.3). PMMA spacers may be used on the back surface of some designs to avoid bottoming out, but may affect the integrity of the cement mantle. Metal backing, initially introduced for more even stress distribution, has been discontinued due to generally poorer results.

The process of bone preparation and cementing technique is crucial to long-term survival. The presence of a radiolucent line in zone 1 of the bone–cement interface on the 12-month radiograph is a predictor of early failure and represents suboptimal technique, rather than a fault in the philosophy of cemented fixation.

The bony bed is prepared by reaming to subchondral bone and the creation of multiple keyholes, increasing the surface area for cement interdigitation to provide fixation and rotational stability. The correct implant size is based on the size of the final reamer used and the preferred cement mantle thickness, following which the flange is trimmed, either by using a template or markings on the flange itself.

Charnley used three major keyholes (ilium, ischium, and pubis), with other cementing proponents advocating the use of multiple smaller ones, or a combination of both techniques.

The bone is washed and dried as thoroughly as possible and the use of a suction vent in the ilium, superior to the rim of the acetabulum, can assist with this. The cement is inserted and pressurized, avoiding laminations, with the cup being inserted at the appropriate time, then held and pressurization maintained until it sets. Any excess cement and osteophytes are removed to avoid impingement.

Initial fixation can be achieved by press-fitting of a hemispherical component, or by the use of a threaded implant, screwed into the host bone.

The use of HA coatings without other surface modifications has now been abandoned, but HA continues to feature in modern designs. Threaded screw-in components were popular in mainland Europe, but have been superseded by the press-fit modular designs. These components may be either hemispherical or non-hemispherical and some have an expanded equator to enhance initial stability.

The press-fit cups are implanted by under-reaming of the acetabulum by 1–2mm, depending upon bone quality and prosthesis design, then the implant is impacted into place. The injudicious use of excessive impaction force must be avoided to reduce the risk of fracture.

If there are concerns regarding the quality of the initial fixation, augmentation can be achieved by additional screw fixation and components come in a variety of designs, some with no screw holes, and some with many (Figure 7.4.3). Screws must be accurately inserted to avoid tilting the prosthesis and distancing it from the surrounding bone. There will be one central hole in a modular cup for the insertion handle, and a cap is provided to occlude this prior to insertion of the liner. Some implants also feature spikes or fins to resist rotational forces, but solid resurfacing cups have no holes and a different design of insertion handle is required.

Modular cups were originally designed for use with polyethylene liners, but were associated with high rates of wear and osteolysis. This was more common with vertical implants (greater than 50 degrees) or when large heads articulated with thin liners. The phenomenon of backside wear was common, when a poor locking mechanism allowed motion to occur between the liners back surface and the inside of the metal shell. This is an example of type 4 wear between two surfaces which are not intended to articulate and the debris produced can access the bony interface via unplugged screw holes, leading to the development of pelvic osteolysis, in keeping with the concept of the effective joint space, as described by Schmalzreid and illustrated by computed tomography studies published by Engh.

Modern implants have improved locking mechanisms, polished inner surfaces, and caps to occlude the screw holes. The modular shells also permit the use of liners with different bearing materials such as highly cross-linked polyethylene, ceramic or metal, but long-term survivorship results of these implants are not yet available.

The development of cemented femoral components has seen the divergence into two main categories, with the establishment of the composite beam (shape-closed) and taper-slip (force-closed) philosophies (Figure 7.4.4).

Composite beam implants have rough surfaces to promote bonding with the cement and may have a collar designed to load the calcar. Once inserted the stem should not move and any debonding or subsidence, at the implant–cement or bone–cement interface, represents loosening and failure.

Taper-slip components are polished, collarless, and taper in two or three planes. There is no bonding with the cement and the stems are designed to subside within the cement mantle, which must therefore be regular, complete, and of the highest quality. Stems are collarless to permit subsidence and polished to avoid abrasion of the inner surface of the cement mantle and the generation of debris, which led to high failure rates in matt finished collarless stems. A hollow centralizer is used to optimize stem positioning, avoid the creation of defects in the cement mantle, promote controlled subsidence, and to eradicate end-bearing, which would lead to distal load transmission.

Subsidence utilizes the viscoelastic properties of cement in a process called creep, which is non-recoverable deformation under load. As the femoral prosthesis is loaded, the cement deforms and the implant subsides within the mantle. The stem subsides and becomes more securely fixed within the mantle, generating hoop stresses within it. These stresses are then transmitted to the surrounding bone, which, because it is being loaded, will not undergo stress shielding, which could lead to the eventual loss of proximal support and loosening. This is in accordance with Woolf’s law and the aim of polished tapered stems is to load the femur as proximally and ‘physiologically’ as possible.

Good long-term results have been achieved with both composite-beam and taper-slip stems, but when the importance of the philosophies or the bone cement are ignored, unacceptably high failure rates can occur as was seen with the Capital hip femoral prosthesis or the use of Boneloc cement.

Cementing technique is crucial and has progressed from the original ‘first generation’ bowl mixing and finger-packing, to vacuum mixing, pulsatile lavage, restrictors, retrograde filling with a gun, pressurization, and the use of centralizers (Table 7.4.1).

The cement mantle should be even, with a minimum thickness of 2–4mm, and the stem in neutral alignment, filling more than half of the femoral canal. Varus alignment of the stem leads to a higher failure rate, but probably reflects a global failure in the surgical technique and implantation rather than in the cemented philosophy.

Results from the Scandinavian joint registries have confirmed improved survivorship with the advent of each generation of cementing technique, confirming that the results of cemented arthroplasties are technique dependant.

There are two main categories of uncemented femoral components depending upon where in the femur the initial press-fit mechanical stability is achieved. The stems can be straight or curved, and feature a variety of porous coatings, beads or fibre-mesh (Figure 7.4.4). The implant may have a collar to prevent subsidence, whilst attempting to load the calcar.

The first category are the stems designed to provide metaphyseal fixation with fit and fill implants, which have porous coating only on the proximal part of the prosthesis. The distal stem may be polished or slotted to prevent distal loading and reduce the incidence of thigh pain.

The proximal coatings on early metaphyseal stems were often incomplete, leading to the development of osteolysis due to wear debris accessing the interface, but modern implants feature circumferential coatings to prevent this and aim to load the femur more ‘physiologically’ via the metaphyseal region.

Because of the success of metaphyseal filling designs and the realization that the distal part of the stem may, in fact, be redundant, shorter or stemless implants, also designed to achieve metaphyseal fixation, have recently entered the marketplace. These stems are often identical to the proximal portion of a design that has been in use for a longer period, but there are no long-term results available, and claims that they will achieve the same results as the stems upon which they have been based can not be supported.

The implants designed to achieve distal diaphyseal fixation are extensively porous coated along all, or most, of the intramedullary surface and may feature longitudinal flutes to enhance rotational stability.

These extensively coated implants obtain the majority of their fixation distally and therefore risk the development of stress shielding, with subsequent proximal femoral bone loss. Because of the extensive coating these stems may also be harder to revise without causing significant additional loss of bone, as a result of which specialized instruments may be necessary for their safe extraction.

Uncemented implants are associated with a greater risk of intraoperative femoral fracture, which if undisplaced and unrecognized can progress to a complete fracture in the early postoperative period. If recognized at the time of surgery the fracture can be stabilized with cerclage wires or cables and the patient kept non-weight bearing for a period. Thigh pain is more commonly reported with uncemented implants, particularly the larger, stiffer, extensively coated ones.

Uncemented femoral components are not recommended for use in patients with a stove-pipe or Dorr type C femur, but such femoral anatomy is also associated with a higher failure rate of cemented implants.

Conventional hybrid fixation consists of a cemented stem with an uncemented cup (Figure 7.4.5). The uncemented modular shells were originally used with a polyethylene liner, but with the growing popularity of alternative bearings, different articulations, and larger head sizes are becoming more popular.

There has also been a recent trend towards the reverse hybrid (uncemented stem and cemented cup) with a metal or ceramic head articulating on a polyethylene bearing.

Good results of hybrid fixation using metal on polyethylene articulations have been demonstrated in the Scandinavian registries, but long-term results of reverse hybrid hips or alternative bearings have not been reported.

The use of a specific implant, whether cemented or uncemented, does not guarantee success. It is the process of implantation, with precise aseptic and surgical technique and the optimal use of the carefully selected components, which will reduce the risk of complications, improving the function and longevity of the arthroplasty.

The results of early uncemented designs were inferior to contemporaneous cemented implants, due to poor implant design and excessive wear of thin polyethylene liners, with many designs being withdrawn, replaced, or becoming obsolete long before any meaningful results could be established.

There is, however, a growing body of evidence from publications and registries that well-designed and well-performed uncemented implants can achieve equivalent results to well-designed and well-performed cemented ones, in most age groups.

The greatest influence on the outcome of any given design of total hip replacement remains the accuracy with which the surgeon implants the prostheses at the time of the operation and the technique of implantation will ultimately take precedence over the technology or specifics of the implant with respect to achieving the best long-term results.