Local Head Roughening as a Factor Contributing to Variability of Total Hip Wear: A Finite Element Analysis

Abstract

Large inter-patient variability in wear rate and wear direction have been a ubiquitous attribute of total hip arthroplasty (THA) cohorts. Since patients at the high end of the wear spectrum are of particular concern for osteolysis and loosening, it is important to understand why some individuals experience wear at a rate far in excess of their cohort average. An established computational model of polyethylene wear was used to test the hypothesis that, other factors being equal, clinically typical variability in regions of localized femoral head roughening could account for much of the variability observed clinically in both wear magnitude and wear direction. The model implemented the Archard abrasive/adhesive wear relationship, which incorporates contact stress, sliding distance, and (implicitly) bearing surface tribology. Systematic trials were conducted to explore the influences of head roughening severity, roughened area size, and roughened area location. The results showed that, given the postulated wear factor elevations, head roughening variability (conservatively) typical of retrieval specimens led to approximately a 30 degrees variation in wear direction, and approximately a 7-fold variation in volumetric wear rate. Since these data show that randomness in head scratching can account for otherwise-difficult-to-explain variations in wear direction and wear rate, third-body debris may be a key factor causing excessive wear in the most problematic subset of the THA population.

About the relevance of roughness parameters used for characterizing worn femoral heads

“…In accordance with conclusions reported in other studies focusing on the effect of hard third bodies on the degradation of total hip prostheses components [3,4,11,18], it is strongly believed that, in the present investigation, in vivo scratching of the titanium-based femoral head is the consequence of a deleterious abrasive third-body wear process due to the centimeter-length metallic fiber.…”

“…As far as the correlation between scratching of femoral heads and the wear of UHMWPE components is concerned, discrepancies have been observed between the quantitative results of different laboratory wear tests and clinical observations [2,3,[6][7][8][9]. These discrepancies have mainly been attributed to the fact that, in clinical situations, localized head scratching may occur and generate damaged regions whose sites, size and severity magnitude are seemingly randomly distributed [2,3,7,10].…”

“…These discrepancies have mainly been attributed to the fact that, in clinical situations, localized head scratching may occur and generate damaged regions whose sites, size and severity magnitude are seemingly randomly distributed [2,3,7,10]. These discrepancies are all the more difficult to interpret that, as claimed by Hall et al [2], considerable work still needs to be undertaken in terms of specifying which relevant roughness parameter should be used in assessing surface texture of femoral heads in relation to the wear of UHMWPE components.…”

“…While the influence of variations in component design, in surgical technique and/or in patient activity level cannot be ignored, the scratching of the metallic femoral head is currently thought to be a major factor affecting the wear rate of the UHMWPE counterface [1][2][3][4][5][6][7][8][9][10]. This scratching phenomenon is increasingly suspected to be caused by hard third bodies such as particles of bone, cement and/or metal, which have often been found embedded in acetabular cups and in the peri-prosthetic tissues [1][2][3][4][5]11,12]. This concern has to be especially taken into consideration when using contemporary constructs with modular interfaces and/or porous metal coatings that are likely to liberate hard metal debris.…”

Four decades of finite element analysis of orthopaedic devices: Where are we now and what are the opportunities?

“…The main focus of this review will be hip and knee replacement, as these have been studied the most, but the issues raised are applicable to the simulation of all orthopaedic devices. FE models have grown in both size and sophistication and techniques exist to assess the initial post-op mechanical environment Zivkovic et al, 2010;Chong et al, 2010;Pettersen et al, 2009;Reggiani et al, 2008;Udofia et al, 2007;Spears et al, 2001;Baldwin et al, 2008;Godest et al, 2002;Taylor and Barrett 2003;Perillo-Marcone and Taylor 2007;Chang et al, 2001) through to the simulation of time dependent processes including bone remodelling induced stress shielding (Huiskes et al, 1987;Perez et al, 2010;Behrens et al, 2009;Gillies et al, 2007;Taylor et al, 2004;McNamara et al, 1997;Weinans et al, 1994;Rietbergen et al, 1993;Gupta et al, 2006), tissue adaptation Andreykiv et al, 2005;Simmons et al, 2001), wear (Strickland et al, 2011;Strickland and Taylor 2009;Knight et al, 2007;Fregly et al, 2005;Bevill et al, 2005;Teoh et al, 2002;Brown et al, 2002;Maxian et al, 1996;Pal et al, 2008), damage accumulation of the cement mantle (Coultrup et al, 2010;Janssen et al, 2009;Lennon et al, 2007;Jeffers et al, 2007;Janssen et al, 2006;…”