Do obesity and/or stripe wear increase ceramic liner fracture risk? An XFEM analysis

Abstract

Background: Hypothesized risk factors for fracture of ceramic liners include impingement, edge-loading, and cup malpositioning. These risk factors are similar to those for generation of stripe wear. However, it is unclear whether the biomechanical conditions contributing to stripe wear generation also increase the risk for ceramic liner fracture

Questions/purposes: We asked whether (1) head stripe wear propensity; and (2) cup orientation would correlate with alumina liner fracture risk for instances of normal and elevated body weight.

Methods: An eXtended Finite Element Method (XFEM) model was developed to investigate these mechanisms. Liner fracture risk for 36-mm alumina bearings was studied by simulating two fracture-prone motions: stooping and squatting. Twenty-five distinct cup orientations were considered with variants of both acetabular inclination and anteversion. Four separate body mass indices were considered: normal (25 kg/m(2)) and three levels of obesity (33, 42, and 50 kg/m(2)). Material properties were modified to simulate alumina with and without the presence of dispersed microflaws. The model was validated by corroboration with two previously published ceramic liner fracture studies.

Results: Of 200 XFEM simulations with flaw-free alumina, fracture occurred in eight instances, all of them involving obesity. Each of these occurred with cups in ≤ 37° inclination and in 0° anteversion. For 200 corresponding simulations with microflawed alumina, fracture propensity was greatest for cups with higher (edge loading-associated) scraping wear. Fracture risk was greatest for cups with lower inclination (average 42° for fractured cases versus 48° for nonfractured cases) and lower anteversion (9° versus 20°).

Conclusions: Fracture propensity for 36-mm liners was elevated for cups with decreased anteversion and/or inclination and under conditions of patient obesity.

Clinical relevance: Factors causing stripe wear, including obesity and cup malpositioning, also involve increased risk of ceramic liner fracture and merit heightened concern.

Fig. 1A–C

The FE model of the global construct consisted of THA hardware and the hip capsule. Two implant geometries were considered: 28-mm (A) and 36-mm (B) head diameters. Fracture risk was greatest for instances of edge-loading caused by deep flexion with or without the occurrence of component impingement. Fracture initiation occurred as a result of development of tensile stress concentrations that exceeded the material tensile strength. These stress concentrations could occur at two distinct sites: the impingement site (as a result of neck-on-liner contact) and the egress site (resulting from femoral head subluxation and resulting edge-loading). In general, the location of the maximum tensile stress in the 28-mm implant occurred near the cup edge, whereas that for the 36-mm implant was located intermediately between the cup edge and cup pole. Modeling fracture initiation for these locations was enabled by enriching the entire head egress region of the cup (C). Additionally, the impingement region was enriched as a separate zone.

Fig. 2

Schematic summary of the 632 individual simulations in the study. Both 28- and 36-mm ceramic implants were considered. For the 28-mm liner, two separate model corroboration series were run, including (1) evaluation of the XFEM model against a LEFM fracture model; and (2) corroboration with an physical simulation of ceramic liner fracture (Please refer to the Appendix for further discussion of model corroboration.). For the 36-mm liner, corroboration was also performed for liner impact scenarios. The stripe wear/fracture series consisted of two fracture-prone maneuvers: stooping and squatting. For each of these two motions, 100 total global FE simulations were run: 25 distinct cup orientations times four values of patient BMI. For each global simulation, two additional XFEM analyses were performed simulating both microflawed and nonmicroflawed alumina. (Please refer to the Appendix for additional discussion of model development.) Fx = fracture.

Fig. 3A–B

The location of edge-loading-associated scraping (stripe) wear generation was similar for the squatting (A) and stooping (B) simulations. Linear wear depth and total volumetric wear were substantially increased for increased BMI. Computed volumetric wear for squatting was approximately double that for stooping.

Fig. 4

Fracture risk correlated with edge-loading-associated volumetric stripe wear computed for both squatting and stooping.

Fig. 5

Cup orientation dependence of fractures encountered for all 200 simulations of microflawed alumina liners. Increased fracture risk was found for liners oriented in decreased anteversion and/or decreased inclination. For cups positioned in 30° of inclination and 0° of anteversion, 87.5% (seven of eight) of the cases resulted in fracture.

Fig. 6

For the 36-mm cups, the crack typically initiated distinctly away from the cup edge and then propagated bidirectionally toward the edge as edge-loading progressively increased with further head subluxation.

Fig. 7A–B

(A) In contrast to the 36-mm cups (Fig. 6), in cases where fractures occurred for the 28-mm implant, fractures initiated always at the cup edge after impingement and subsequent edge-loading. As the femoral component continued further in flexion, edge-loading severity increased, leading to the crack then propagating away from the cup edge. The location of fracture initiation was similar to that determined from LEFM fracture analysis of an identical bearing (B) [11].

Fig. A1A–D

Schematic of the XFEM numerical formulation. In standard FE modeling (A), the displacement field must be a continuous function across any given element. To model a discontinuity within a given solid object (B), a conventional mesh must be structured such that the discontinuity lies across the element boundaries. However, XFEM (C) allows for mesh-independent modeling of discontinuities by incorporating enrichment features to augment the standard FE displacement approximation. Element enrichment functions (D) allow for discontinuities to exist within a given element, and they allow for approximating stress singularities (unboundedly accentuated local stress concentrations) in the near neighborhood of crack tips. In these mathematical expressions, u_i is the usual nodal displacement vector, N_i(x,y) represents the usual nodal shape functions, a_i and b_i are enriched degree-of-freedom vectors, H(x,y) is the Heaviside step function, and F_α(x,y) are crack tip functions.

Fig. A2A–B

An XFEM corroboration series was conducted by computationally replicating the neck-on-liner impact fracture scenario of a previously reported experimental investigation (A; reprinted from Maher SA, Lipman JD, Curley LJ, Gilchrist M, Wright TM. Mechanical performance of ceramic acetabular liners under impact conditions. J Arthroplasty. 2003;18:936-941, with permission from Elsevier). This corroboration study considered both 36-mm (B) and 28-mm (A) ceramic liners. Under displacement control, the femoral neck was displaced along an axis passing through the center of the neck and the site of neck-on-liner impingement encountered during a stooping maneuver for cups positioned in 45° inclination and 0° anteversion.

Fig. A3

LEFM/XFEM corroboration series. XFEM fracture analysis of 28-mm bearings demonstrated similar dependency on cup inclination as that for an LEFM formulation. For both XFEM and LEFM, fracture of the liner occurred for cups positioned in ≥ 35° of inclination for 10° of anteversion. (The criterion for fracture in the LEFM series was that the K_I stress intensity factor exceeded the critical stress intensity factor, K_Ic.) Data reprinted from Elkins JM, Pedersen DR, Callaghan JJ, Brown TD. Fracture propagation propensity of ceramic liners during impingement-subluxation. J Arthroplasty. 2012;27:520-526, with permission from Elsevier.

Fig. A4A–B

Neck-on-liner impact fracture XFEM corroboration from a physical experiment [20]. For both 28-mm (A) and 36-mm (B) liners, several simulations were run by increasing the amount the neck displacement (leading to increased impact force) into the ceramic liner (see Fig. A2). A fracture threshold contact force of 23.2 kN was determined for 28-mm liners (A). A similar threshold force of 24.5 kN was determined for the 36-mm liners (B). These compare favorably with an experimentally determined threshold of 23 kN for alumina ceramic liners [20].