Contribution of Micro-Motion to Backside Wear in a Fixed Bearing Total Knee Arthroplasty Rayna A.C. Levine, Kathleen A. Lewicki, John H. Currier, Michael B. Mayor, Douglas W. Van Citters Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire 03755 Received 28 July 2015; accepted 17 February 2016 Published online 16 March 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.23203 ABSTRACT: This study seeks to identify important factors related to backside wear of tibial inserts in vivo and determine an appropriate wear model for backside wear. An IRB approved database was queried for tibial inserts of a single design from one manufacturer that exhibited evidence of rotatory motion on the backside of the polyethylene. These devices were measured for volumetric wear using a previously established protocol. Features including the change in locking lip width and measurement of micro- motion marks were used to describe the motion pattern. Volumetric wear and implant characteristics were compared using linear regressions by modeling wear theories suggested by Archard and Wang to determine the most appropriate model for backside wear. The Wang model showed that duration, adjusted sliding distance, and cross-shear index accounted for approximately 58% of the volumetric wear variation while adjusted sliding distance and duration in vivo accounted for approximately 35% of the volumetric wear variation in the Archard model. Patient weight (p ¼ 0.750), patient BMI (p ¼ 0.680), and backside area (p ¼ 0.784) of the tibial insert were all found to be non-significant in the Wang model. Similarly, patient weight (p ¼ 0.233), patient BMI (p ¼ 0.162), and backside area (p ¼ 0.796) were found to be non-significant in the Archard model. Multidirectional micro-motion appears to contribute significantly to the wear of these components, supporting the Wang theory of cross-shear for polyethylene wear. Cross-shear of polymers on an unpolished titanium tray can lead to an increase in wear debris in the body. Care should be taken when designing locking mechanisms and tray designs. ß 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 34:1933–1940, 2016. Keywords: polyethylene; wear; total knee arthroplasty (TKA); micro-motion Total knee replacement is a highly successful treat- ment for patients suffering from severe osteoarthritis, other degenerative joint disease, and many other causes of knee pain. Modern arthroplasty devices replace the cartilage of the knee with an ultra-high molecular weight polyethylene bearing, which articu- lates with a femoral component often made of cobalt- chromium-molybdenum (CoCrMo) alloy. On the tibial side, the polyethylene tibial insert is often supported by a metal tray, which may be a CoCrMo alloy or a titanium alloy. The polyethylene is often secured to the tray by one of a variety of proprietary locking mechanisms, comprising a fixed-bearing device. A frequently cited reason for failure of these knee replacements is wear of articular and/or backside (non-articular) surface of this polyethylene tibial insert. 1–4 For knee replacement patients, the sub-micron polyethylene debris produced by wear of the tibial insert can cause a number of negative outcomes in- cluding synovitis, aseptic loosening, and osteolysis. 2,4–6 Although in fixed-bearing tibial constructs the polyeth- ylene is captured by the metal tray, previous studies have indicated that the backside surface of the tibial insert can move relative to the tray, and this micro- motion is a potential source of significant debris. 7–11 In the interest of designing better knee replacement products, understanding the factors contributing to backside wear—including both patient and design factors—is fundamental. To better understand and predict design factors contributing to backside wear, it is useful to accurately model the wear interaction between the polyethylene tibial insert and the metal tibial tray counter-face. Although Archard’s model 12 is a good first approxima- tion for sliding wear and holds true for many bearing couples, it specifies that load is a major determinant of wear rate. Wear testing studies have shown that load does not appear to have a significant effect on wear rate in polyethylene, 13–15 and even Archard’s manu- script suggests that his wear model may not apply to all combinations of materials, or to all materials under all conditions. 12 This is particularly true with poly- mers because Archard’s model depends on local yield- ing of surface asperities as might happen in metals, whereas lower yield-stress materials likely experience regional yielding in response to the applied contact conditions. In contrast to Archard’s model, models derived by Hippensteel et al. 13 and Mazzucco and Spector 14 find that contact surface area is significant in determining wear, but load is not. Beyond contact geometry, crossing motion has also been shown to be significant contributor to polymer wear. 15 Wang suggests that this may be due to the ability of non-cross-linked polyethylene lamellae to align with the primary direc- tion of motion, making them more susceptible to wear in multidirectional motion. 15 The general tenets of Wang’s model have proven durable over time with additional consideration given to crossing intensity and rolling-sliding conditions. 16,17 While this has been applied to laboratory models and in vivo wear studies of the intended articulation in total knee RACL and KAL contributed equally and are co-first authors. Grant sponsor: National Science Foundation Graduate Research Fellowship; Grant number: No. DGE-1313911. Correspondence to: Douglas W. Van Citters, (T: (603) 646-6406; F: (603) 646-3856; E-mail: dvancitters@dartmouth.edu) # 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. JOURNAL OF ORTHOPAEDIC RESEARCH NOVEMBER 2016 1933