Introduction

Cementless fixation in unicompartmental knee replacement (UKR) is a developing field with advances in implant design and fixation surfaces. Although cementless options in unicompartmental replacement have been available for nearly 20 years, the operation has not become widespread. Cemented fixation is used in the majority of UKR procedures, with good clinical results. However, there is a concern over the suitability of cemented fixation in the younger, high-demand patient. Cementing in UKR is difficult and cementing errors, such as cement loose bodies, are common and may cause failure. In addition, the slow development and general acceptance of cementless UKR may be in part due to poor results from early designs. The use of cementless fixation in UKR is likely to increase with the aim for long-term biologic fixation being a realistic goal.

Rationale

The development of cementless UKR has mirrored that of total knee replacement (TKR), with the use of porous coatings, with and without hydroxyapatite, and, in some designs, the addition of cancellous screw augmentation. All designs have utilized a keel on the tibial component, although the shape, size, and direction have shown considerable variation (Figs. 12–1 and 12–2). The femoral component has usually mirrored the cemented equivalent for each design, although those femoral designs with a single peg had a second peg added for rotational stability.

Figure 12–1 A cementless Oxford tibial component demonstrating a porous titanium layer with a hydroxyapatite coating.

Figure 12–2 A cementless, hydroxyapatite-coated Unix tibial component with a horizontal keel and screw holes for additional fixation.

(Photograph courtesy of Jean-Alain Epinette.)

One of the first advocates of cementless UKR was Jean-Alain Epinette,¹ who designed and developed a hydroxyapatite-coated tibial component. This component differed from most designs by employing a horizontal keel that slotted beneath the tibial spines, thus maintaining bone stock beneath the tibial tray. In addition, there were cancellous screws to augment the initial fixation. Kaiser and Whiteside further highlighted the importance of initial stability in 1990, with a cadaveric in vitro study demonstrating the additional stability provided by cancellous screws in a unicompartmental tibial component, compared with a posteriorly angled peg.²

Cementless fixation has many potential advantages over the more commonly used cemented fixation. In hip arthroplasty, cementless fixation has become ever more popular, whereas cementless total knee arthroplasty has remained less popular, particularly with regard to the tibial component. This is probably due to the increased difficulty in obtaining good initial fixation with a mostly flat tibial component resting on a flat tibial plateau. Although cementation provides an adequate fixation method in most cases, the use of cement results in several consequences that encourage perseverance with the development of cementless fixation. The disadvantages of cement use include possible thermal damage to the bony surfaces, an increased operative time, and the risk of cementing errors such as cement loose bodies or incorrect cement loose bodies, incorrect seating or impingement on cement. Cementless fixation avoids these problems and has the advantage of preserving tibial bone, as well as the theoretical potential of very long-term biologic fixation. However, cementless fixation in total knee arthroplasty has struggled to gain widespread acceptance due to some poor results, with a particular risk of failure of ingrowth.

Cementless fixation in UKR has mainly been used in the few centers that are strong advocates of unicompartmental surgery. The latest reports from the Australian and New Zealand joint registries show that cementless UKR accounts for 11% and 6% of all UKR procedures, respectively.^3,4 The Swedish register reported that all the UKRs performed in 2009 had both components cemented.⁵ Neither the National Joint Registry of England and Wales nor the Canadian registry gives information regarding fixation method for UKR.^6,7

Although the presence of thin, nonprogressive, radiolucent lines (Fig. 12–3) has long been recognized in knee arthroplasty, with no associated increase in failure rate, the exact cause and effect of these lines is still unknown. There is, however, a perception that a radiolucent line is indicative of suboptimal fixation. The incidence of radiolucency differs beneath different implants, suggesting that the mechanical environment is important in their development. Forsythe et al.⁸ reported an incidence of over 50% with the cementless Whiteside Ortholoc II UKR, whereas Pandit et al.⁹ reported just 7% of cementless Oxford tibial components had a partial radiolucency, with none with a full radiolucency. There is also a reported increase in incidence of radiolucent lines with cemented Oxford components compared to the cementless version.⁹ There is no clear consensus on the role or importance of thin (≤1 mm), stable, so-called physiologic radiolucent lines, although there is evidence that they are not associated with either loosening or a decrease in clinical outcome scores. However, the greatly reduced incidence in cementless UKR is encouraging and suggests good bony fixation (Fig. 12–4). The Oxford UKR is particularly suited to cementless fixation because of the fully mobile bearing. The bearing movement causes a marked reduction in shear forces being transmitted through the tibial tray, and therefore the tray is mostly subjected to compression. This is an almost ideal mechanical environment for cementless fixation. Radiostereometric analysis studies of cementless fixation in TKR have shown a consistent pattern of migration, which differs from cemented fixation. Cemented components usually demonstrate continuous early migration, but the magnitude is small and stability is achieved by 2 years. In contrast, cementless components usually migrate a larger amount in the first few months before stabilizing. Onsten and Carlsson have demonstrated a reduction in the movement between 1 and 2 years postoperatively with the addition of a layer of hydroxyapatite to a porous-coated PFC tibial component.^10,11 Likewise, Regner et al. demonstrated the reduction in maximum total point movement at 5 years in the Freeman-Samuelson TKR tibial component with the addition of hydroxyapatite to the undersurface.¹²

Figure 12–3 A physiologic radiolucency beneath a cemented Oxford tibial component.

Figure 12–4 A well-seated cementless Oxford tibial component demonstrating no radiolucency at 5 years.

Indications

The indications for UKR are controversial and vary according to the surgeon’s philosophy. At Oxford, our view is that there should be full-thickness cartilage loss on both the medial tibial plateau and femoral condyle (bone-on-bone osteoarthritis), any intra-articular varus deformity should be correctable, and the anterior cruciate ligament (ACL) should be intact. Patello-femoral joint damage is ignored unless there is severe damage laterally. Cartilage damage on the lateral femoral condyle is permissible provided there is full-thickness cartilage present on a valgus stress radiograph and there is no full-thickness loss centrally. Full-thickness damage on the medial side of the lateral femoral condyle is not a contraindication.¹³ We do not consider age, gender, activity level, or Body Mass Index to be contraindications. These indications are our routine for both cemented and cementless UKR. Bontemps also noted that the indications for cemented and cementless UKR are the same; however, due to cost saving, he favors cemented fixation in the elderly.¹⁴ The suitability of cementless UKR in cases of osteonecrosis or areas of bone loss is less easily described. The consensus in our group is that, if the knee is suitable for cemented UKR once the bony surfaces are prepared, then cementless UKR is also indicated. Our rationale for this is based on evidence that one does not require total bony ingrowth to maintain implant stability and therefore, if the bony surface can support the implant, then either fixation method is acceptable. Epinette’s group demonstrated implant stability in cementless UKR when the amount of bony contact between tibia and component was between 38% and 52%¹ (Fig. 12–5).

Figure 12–5 A specimen obtained at revision (for lateral compartment progression) 7 years after initial implantation, demonstrating bony apposition to the hydroxyapatite-coated undersurface of a Unix tibial component.

(Photograph courtesy of Jean-Alain Epinette.)

Setup And Equipment

As with all arthroplasty surgery, the patient setup is of paramount importance. The patient should be set up as for routine medial unicompartmental knee arthroplasty. We recommend the use of a secure leg support, placed at midthigh, to allow the leg to hang and flex fully during surgery and to avoid pressure on the popliteal structures when fully flexed (Fig. 12–6). The Oxford cementless UKR uses the standard equipment trays in addition to the cementless tray. The specialist cementless tray includes instruments and trials specific for the cementless procedure. It contains cementless femoral trials, tibial templates with narrower keel slots, a narrower keel slot pick, and a cementless tibial component introducer. It is essential to use a narrower keel slot saw blade (Synvasive, Reno, NV). The keel slot width is of paramount importance, and therefore particular care needs to be taken when cutting the slot. The narrow keel slot pick also allows clearance of the keel slot without widening it unnecessarily. While the femoral component does not require any special equipment, implantation of the tibial component requires a specific introducer (Fig. 12–7). The introducer for implanting the Oxford tibial tray allows accurate placement of the tray, ensuring it is placed far enough back on the plateau before release and subsequent final seating. This facilitates firm seating on the posterior cortex and prevents anterior overhang.

Figure 12–6 The correct setup for cementless Oxford UKR. The leg is able to hang vertically and can be flexed easily during surgery. The thigh support is midthigh so popliteal fossa compression does not occur.

Figure 12–7 The Oxford cementless tibial tray attached to the specially designed introducer. The component can be accurately positioned and partially impacted before removal of the introducer to allow clearance of soft tissue beneath the tray and final impaction.

Approach

Initially the Oxford UKR was implanted through a long midline incision as used for TKR. The Phase 3 instrumentation was introduced to allow the use of a short medial incision. Clinical studies have demonstrated that patients make a quicker recovery postoperatively and have equal outcome scores.^15–18 Therefore, the Oxford cementless UKR utilizes a short medial incision, from the medial pole of the patella to the tibial tubercle, as per the standard approach for cemented arthroplasty. Although there are advantages to a shorter incision, adequate exposure is important and the incision must be long enough to allow adequate inspection of the other compartments, retraction of the patella, and excision of osteophytes. The anteromedial corner of the tibia should be exposed to allow seating of the tibial cutting jig and exposure, and subsequent removal, of any osteophytes. Ligaments are never released, and care is taken to protect the medial collateral ligament (MCL). A specific retractor has been designed to help protect the MCL. The patella should not be inverted but can be retracted laterally. Introduction of a femoral intramedullary rod is helpful for alignment of the femoral component, as well as acting as an effective patellar retractor. Once the decision to proceed with UKR has been made, all osteophytes around the medial femoral condyle, intracondylar notch, and anterior aspect of the medial tibial plateau should be removed. There is often an osteophyte anterior to the insertion of the ACL.

Technical Description

Careful and accurate preparation of the bony surfaces is of great importance as good initial fixation is reliant on a tight press-fit interface between both the femoral and tibial components and host bone. Different implant designs require different techniques. The following is a summary of the Oxford technique. An extramedullary tibial guide is used for tibial preparation. It should be positioned so the horizontal cut is just beneath the tibial defect unless this is very deep. The vertical tibial saw cut is made using a reciprocating saw (Synvasive, Reno, NV) placed just medial to the apex of the medial tibial spine and should be directed toward the hip joint or anterior superior iliac spine. The saw should be parallel to the tibial guide and should stop about 1 mm above the guide so as not to damage the bone supporting the implant. The horizontal cut is made with a 7° slope, taking care not to damage the MCL. Following removal of posteromedial femoral osteophytes, the size of the flexion gap should be checked to ensure that it is large enough for the minimum tibial component and bearing. If not large enough, more bone should be removed from the tibia. The position of the femoral component is determined using a femoral drill guide. Ideally the component should be flush with the original joint surface posteriorly and central on the femoral condyle. Bone is removed progressively from the distal femur until the flexion and extension gaps are balanced. The extension gap should be assessed in 20° of flexion so that the posterior capsule is lax. A menisectomy is performed. Ideally, a small rim should be preserved medially to protect the MCL from the implant. Bone and cartilage are removed anterior and posterior to the implant to prevent impingement. A slot for the tibial keel is fashioned. A trial reduction is undertaken with the definitive trial components to ensure they seat fully, do not impinge, and function satisfactorily. The definitive implants are then impacted until secure.

Postsurgical Follow-Up

We use our standard postoperative rehabilitation regimen for both cemented and cementless UKR. Patients are allowed to weight-bear once comfortable. The patient is discharged once he or she is mobile on crutches and can safely negotiate stairs. In our institution, there is no difference in time to discharge between cemented and cementless UKR. Patients are advised to mobilize as comfort allows and are warned that it may take many months for the soreness, stiffness, and swelling to settle.

Before discharge, patients have routine radiographs with the anteroposterior view taken parallel to the tibial tray, allowing assessment of the component-bone interface. The lateral view is perpendicular to the femoral component, allowing assessment of bony contact around each peg. Using carefully screened radiographs enables proper evaluation of the interfaces and allows comparison with subsequent radiographs. The accuracy of detecting radiolucent lines, and hence assessment of whether they are progressive, is dependent on the reproducible acquisition of radiographs. The incidence of radiolucent lines beneath the cemented tibial component in Oxford UKR is approximately 62%.¹⁹ However, the incidence in cementless Oxford UKR is markedly reduced at 1 year, with only 7% having a partial radiolucency and none having a complete radiolucency.⁹ Radiographs taken in the first few days after surgery sometimes demonstrate a thin radiolucent line beneath the tibial tray (Fig. 12–8A). This is indicative of the tray not being fully seated at the time of surgery. However, with normal postoperative rehabilitation the radiolucent line disappears and the tray becomes securely seated on the tibial plateau (Fig. 12–8B).

Figure 12–8 (A) A postoperative radiograph of a cementless Oxford UKR demonstrating a radiolucency beneath the tibial component. (B) A radiograph at 1 year of the same Oxford cementless UKR showing stabilization of the tibial component and no radiolucency.

(Reproduced with permission and copyright © of the British Editorial Society of Bone and Joint Surgery from Pandit H, Jenkins C, Beard DJ, et al. Cementless Oxford unicompartmental knee replacement shows reduced radiolucency at one year. J Bone Joint Surg [Br] 2009;91:185-189. [Figures 3a and 3b].)

Clinical Results

Clinical results of cementless UKR are not commonly reported, and those reports evident in the literature are usually short series. A single randomized, controlled trial of cemented versus cementless UKR at Oxford has shown no difference in clinical outcome, and a reduced incidence of radiolucency with cementless fixation, at 2 years.²⁰ There have also been reports that contribute to the base of knowledge in UKR fixation, with lessons being learned and improvements made. A report from Keblish and Briard in 2004 regarding the LCS mobile-bearing UKR highlights the reasonable clinical results obtainable, but again shows that there are concerns regarding cementless fixation as all the failures of fixation in the study were in the cementless group.²¹ The Porous Coated Anatomic UKR had initial success, using both cemented and cementless fixation methods. However, Bernasek et al. described an increased number of failures due to fixation in the early postoperative period, with fibrous tissue forming at the implant-bone interface²² in the cementless components. Lindstrand and Stenstrom reported good results initially with the same implant, but then showed a deterioration between 4 and 8 years postoperatively.^23,24 Although the failures described in the midterm were mostly related to extreme polyethylene wear, this demonstrates the difficulty in obtaining long-term follow-up on UKR fixed without cement. Epinette and Manley have provided perhaps the best mid- to long-term follow-up of a cementless UKR.²⁵ At a minimum follow-up of 5 years, with a maximum of 13 years, there were no failures of fixation from 125 hydroxyapatite-coated Unix knee replacements. Radiographs obtained 14 years after surgery show that excellent fixation without the development of radiolucencies can be achieved (Fig. 12–9). The results of the available published reports are summarized in Table 12–1. The data presented demonstrate the need for further studies with longer follow-up in addition to focus on particular patient groups, such as the young with high demand.

Figure 12–9 Anteroposterior (A) and lateral (B) radiographs taken 14 years postoperatively in a female patient who had a medial UKR for osteoarthritis at the age of 72. Note the absence of radiolucency either beneath the tibial component or around the screws.

(Radiograph courtesy of Jean-Alain Epinette.)

Summary

At present, cementless UKR is still an unproven long-term treatment for single-compartment osteoarthritis. However, the majority of reported failures are due to either inherent problems with design or progression in another compartment, rather than the fixation method. The important factors for success in cementless UKR surgery are the same as for any arthroplasty surgery: adherence to appropriate surgical indications, use of a well-designed implant, and meticulous surgical technique. Although due regard for the short- to midterm follow-up must be made, recent results show an encouraging future for cementless UKR.