Failure of Primary Total Knee Arthroplasty: Establishing a Baseline for Retrieval Analysis Using Well-Functioning Necropsy Specimens
Date of Award
Master of Science (MS)
William M. Mihalko, MD, PhD
Richard A. Smith, PhD John L. Williams, PhD
Failure mechanisms, Inflammatory cytokines, Laxity, Tissue metal content, Total knee arthroplasty, Wear
Introduction. Total knee arthroplasty (TKA) is an effective treatment for end-stage osteoarthritis of the knee. While many patients have positive functional and pain outcomes following primary TKA, a subset of patients have suboptimal outcomes, such as unexplained pain or aseptic loosening of the implant components necessitating revision of the index procedure. The understanding of the exact etiology of these suboptimal outcomes of primary TKA is in its infancy. In order to elucidate the etiology of failure in primary TKAs, a baseline for the factors contributing to arthroplasty failure and dissatisfaction must first be established. The purpose of this study was to investigate the relationship between soft tissue laxity, inflammatory cytokine concentrations, tissue metal concentrations, and wear scores of well-functioning implants retrieved from cadaveric specimens to determine the role of each of these factors in implant survivorship. It was hypothesized that decreased joint laxity would increase metal concentrations in the periprosthetic tissue, that cytokines specifically tied to the innate immune system would be elevated in laxer joints, and, lastly, that elevated inflammatory cytokines in the synovial fluid would be associated with elevated periprosthetic tissue metal concentrations.
Methodology. A total of 33 cadaveric specimens with primary TKA were obtained from two sources, the Medical Education and Research Institute (Memphis, TN) and RestoreLife USA (Elizabethton, TN), as part of institution review board approved multi-institutional orthopaedic retrieval program. Prior to testing and retrieval, fluoroscopically-assisted radiographs were taken to assess if any evidence of radiolucencies were present and all replacements were determined to be well-fixed per the images. Synovial fluid was then aspirated from the joint, processed, and stored in a -80˚C freezer. Each specimen was mounted into a custom knee testing platform and the IE rotation, varus-valgus (VV) deflection, and AP translation was measured at 0˚, 30˚, 60˚, and 90˚ of flexion. After collection of the laxity data, tissue samples were collected from the supra- and infra-patellar regions, the medial and lateral gutter, and from the tibia. The polyethylene inserts were assessed for wear on the condylar, backside, and stabilizing post surfaces, when applicable, and the femoral condyle was assessed for damage. A bead-based multiplex assay using the Luminex MAGPIX platform (R&D Systems, Minneapolis, MN) was performed on the synovial aspirates for simultaneous detection of various inflammatory cytokines including: IL-1β, IL-6, MCP-1, and MIP-3α. Inductively coupled plasma mass spectrometry (ICP-MS) was performed at Brooks Applied Labs (Bothel, WA) on the periprosthetic tissue samples for determination of tissue cobalt (Co), chromium (Cr), and titanium (Ti) concentrations. Finally, statistical analysis was conducted using SigmaPlot (Systat Software, Chicago, IL) to elucidate whether any correlations existed between the aforementioned factors.
Results. Decreased IE laxity at full extension was inversely correlated with increased Co concentrations in the periprosthetic tissues (ρ=-0.64, p=0.02) with a sample size (n) of 13 and a 95% confidence interval (CI) from -0.88 to -0.14. At 90° of flexion, anterior laxity was inversely correlated with Co concentrations (ρ=-0.66, p=0.03, n=11, 95% CI: -0.90 vi to -0.10). At 60° of flexion, anterior laxity was inversely correlated with Cr concentrations (ρ=-0.63, p=0.03, n=13, 95% CI: -0.88 to -0.12). Anterior laxity at 90° of flexion was also inversely correlated with Cr concentrations (ρ=-0.81, p=0.003, n=11, 95% CI: -0.95 to -0.41). Lastly, posterior laxity at 90° of flexion was inversely correlated with Cr concentrations in the periprosthetic tissues (ρ=-0.74, p=0.01, n=11, 95% CI: -0.93 to -0.25). At 60° of flexion, IE rotational laxity was inversely correlated with IL-1β concentrations (ρ=-0.55, p=0.02, n=18, 95% CI: -0.81 to -0.11). At 60° of flexion, VV laxity was inversely correlated with TNF-α concentrations (ρ=-0.78, p=2x10-4 , n=14, 95% CI -0.93 to -0.43). Additionally, VV laxity was inversely correlated with IL-1β concentrations (ρ=-0.48, p=0.04, n=18, 95% CI: -0.77 to -0.02) at 60° of flexion. At full extension, VV laxity was inversely correlated with IL-6 concentrations (ρ=-0.46, p=0.04, n=19, 95% CI: -0.76 to -0.01). Anterior laxity was directly correlated with IL-6 concentrations (ρ=0.53, p=0.02, n=18, 95% CI: 0.08 to 0.80) at 90° of flexion. Additionally, anterior laxity was directly correlated with MCP-1 concentrations (ρ=0.62, p=0.006, n=18, 95% CI: 0.22 to 0.84) at 90° of flexion. Cr was inversely correlated with IL-6 (ρ=-0.52, p=0.01, n=21, 95% CI: -0.78 to -0.11). Cr was also inversely correlated with MIP-3α (ρ=-0.46, p=0.04, n=21, 95% CI: -0.74 to -0.04).
Discussion. The specimens included in this study consisted of primary TKA implants that were retrieved at necropsy and were determined to be well-fixed per fluoroscopic analysis. The first objective of this study was to establish a baseline of different factors that likely contribute to failure of primary total knee replacements. This study provided semi-quantitative assessment of wear of the condylar surface, backside surface, and stabilizing post, when applicable, and of damage to the bearing surface of the femoral components. Joint laxity was measured using a custom knee testing platform that had been previously studied and utilized in published journal articles and theses. Inflammatory cytokine profiles in the synovial fluid samples were presented and measureable levels of metal debris was found in the periprosthetic tissue samples. The second objective of this study was to determine if any significant relationships arose in correlation analysis of the latter three aforementioned factors. An inverse trend was observed between joint laxity and tissue metal concentrations, such that decreased laxity induced metal release from the implant components. A direct trend was observed specifically between increased anteroposterior laxity and elevated inflammatory cytokines. Lastly, an inverse trend was found between decreased tissue metal concentrations and increased inflammatory cytokines. The third, and future, objective of this study is to utilize these measurements and observations for comparison with failed implants retrieved at time of revision. While some meaningful relationships were observed in this study, there were several limitations that must be noted. Firstly, this study had a relatively small sample size. Secondly, the cohort included a wide range of implants including cruciate-retaining, posterior-stabilized, fixed bearing, mobile bearing, monoblock, and modular designs fabricated from different materials
Perkins, Meredith (http://orcid.org/0000-0002-6492-0784), "Failure of Primary Total Knee Arthroplasty: Establishing a Baseline for Retrieval Analysis Using Well-Functioning Necropsy Specimens" (2018). Theses and Dissertations (ETD). Paper 460. http://dx.doi.org/10.21007/etd.cghs.2018.0451.
Investigative Techniques Commons, Other Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons, Surgical Procedures, Operative Commons, Therapeutics Commons
Joint program with the University of Memphis.