This is a prospective randomized control trial that aims to investigate fixation of implants after revision total knee replacement. The investigators are comparing long-stemmed revision implants with hybrid cementation to short-stemmed primary implants with augment using a special type of x-ray imaging (radiostereometric analysis) to assess implant movement over time. The investigators are also comparing costs and patient-reported outcomes between the groups.

The demand for total knee replacement (TKR) continues to grow, especially for younger patients who are at a greater risk of outliving the useful lifetime of their implants, thus requiring revision surgery (Aggarwal et al., 2014; Stambough et al., 2014). Rates of revision TKR are growing at an even faster rate than primary TKR. Between 2009-10, the incidence of primary TKR grew by 9.4% in the U.S., while revision TKR grew by 17.1% (Kurtz et al., 2014). Revision occurs for failures such as loosening, infection, or wear. Unfortunately, patients undergoing revision TKR are roughly 6 times more likely to require re-revision compared to patients undergoing primary TKR, further increasing the revision burden. Standard of care for revision TKR has evolved, in search of long-lasting fixation and improved patient outcomes. In many centres, the current standard of care is to use revision TKR implants with long stems for enhanced fixation. Surgeon preference is variable, however, as to whether these stems will be fully cemented, or whether only the most proximal aspect of the baseplate will be cemented (hybrid fixation). While most common, these long-stemmed revision implants have suffered from poor survivorship, are technically demanding to implant, and can result in poor patient outcomes, including ongoing pain. Over the past decade porous metal augments have gained favour. These devices interface between the defective bone and the implant, aiding fixation. The first augment of its kind are the Trabecular Metal cone augments from Zimmer. Although an improvement in principle, they have a number of drawbacks including poor fit with most implant systems, increasing technical challenge and requiring intra-operative customization, which lengthens the surgery. More recently, Stryker Tritanium augments have been introduced that offer a closer fit with Triathlon revision implants, decreasing operative time. This closeness of fit also provides an opportunity to rethink the utilization of long stems when porous augments are used. Ultimately, the goal would be for porous metal augments to completely fill all bone defects, essentially enabling a more primary-like TKR to be performed, enhancing patient outcomes. The gold standard for measurement of implant fixation over time is radiostereometric analysis (RSA). At the time of surgery, 0.8-1.0 mm Tantalum beads (typically n = 6-8 per region of interest) are inserted into the bone surrounding the implant of interest, and occasionally also into the polyethylene liner or insert of the implant. RSA exams are then acquired within the first 2 weeks post-operation, and again at 6 weeks, 3 months, 6 months, 1 year, and 2 years. Each exam involves a stereo x-ray acquisition with a calibration unit to reconstruct the 3D location of the implant relative to the marker beads. Across each exam, the migration of the implant relative to the marker beads is tracked compared to the baseline exam acquired within the first 2 weeks. RSA is extremely precise and accurate, and is able to track migrations on the order of 10's of microns. Large early migrations within the first 1-2 years post-operation are predictive of eventual implant loosening and failure. RSA can also be used to measure inducible displacement of implants to assess instantaneous fixation of implanted components. These RSA exams compare a supine, unloaded image of the patient's joint to a standing, loaded image of their joint, to determine whether any movement occurs between the implant and marker beads with loading. To our knowledge, only three published RSA studies of revision TKR exist in the literature. Jensen et al. (2012) published a 2-year study with 40 patients comparing cement reconstruction to Trabecular Metal cones. They found no significant difference between groups for migration at 2 years, and determined that use of the Trabecular Metal cones resulted in better early stability with reduced irregular motion patterns compared to using hybrid cement reconstruction. Heesterbeek et al. (2016) published a 2-year study with 32 patients that compared hybrid to fully cemented stem fixation using the Legion revision system, which was followed up to 6.5 years by Kosse et al. (2017). At both follow-up time points, there was no difference in overall migration between the hybrid and fully cemented stem groups. Taken together, these studies do not reflect the latest thinking in revision knee surgery, do not account for the current generation of augment design, do not directly compare augments with hybrid and fully cemented fixation, and do not correlate fixation and patient outcome scores.

Inclusion Criteria:

• Patients undergoing a first total knee revision surgery

Exclusion Criteria:

• Patients who have previously undergone revision knee surgery
• Patients that require a hinged revision implant
• Patients that do not speak or understand English (questionnaires provided in English)
• Patients with any active or suspected latent infection in or about the knee joint
• Patients with bone stock compromised by disease, infection or prior implantation which cannot provide adequate support and/or fixation to the prosthesis
• Patients that live >100km from our centre (as such individuals are less likely to be available for the required follow-up appointments)

Patients will be blinded so that they do not know what type of implant construct they received. It will not be possible to blind the surgeon or the individual performing the RSA image processing.