The Designs
Testing was performed on two commercially available acetabular shell systems that typified contemporary design concepts and on one laboratory control shell. All three shell systems were reliant on a taper lock as the sole means of liner retention. One of the commercial designs, the Equator Plus (Portland Orthopaedics, St. Clair, Michigan), made use of a cobalt-chromium-molybdenum banded alumina liner surrounded by a 4-mm-thick titanium alloy (Ti-6Al-4V) shell with an apical hole and three fixation screw holes (Fig. 1-A). The other commercial design, the LINEAGE (Wright Medical Technology, Arlington, Tennessee), employed an alumina liner with a 6.5-mm-thick titanium alloy (Ti-6Al-4V) shell and a single apical hole (Fig. 1-B). The laboratory control design had an all-alumina liner with a 4.5-mm-thick titanium shell and a single apical hole (Fig. 1-C).
The Analog Modeling Method
Squire et al.4 reported in a clinical series that nineteen of twenty-one underreamed and impacted shells of a single design exhibited a mean measurable diametric deformation and standard error of 0.16 ± 0.16 mm. They correlated this to between 0 and 1539 N of compression, with the variance attributed to bone quality. In order to reduce this variability in the current evaluation, a pilot study was devised to impact a shell into an underreamed analog pelvis manufactured with a fiberglass cortex surrounding a rigid polyurethane foam cancellous core (Pacific Research Laboratories, Vashon, Washington). This captured the general material properties of bone as well as the geometric structure of the pelvis. A 4.5-mm-thick titanium spherical shell without holes and with an outside diameter of 59 mm was impacted into an analog pelvis that had been reamed to 58 mm, and the shell deformation was measured with use of laser profilometry (Hawk 3D laser scanner, model 5-4-4; Nextec Technologies 2001, Tirat HaCarmel, Israel) and a dial bore gauge (Mitutoyo, Kawasaki, Kanagawa, Japan). The maximum measured deformation in the analog pelvis was 0.12 mm. This was a near-perfect fit with the data obtained from a dynamically loaded finite-element model of the same shell under two-point compression (Fig. 2). When the spherical control shells were placed under mechanical two-point loading, 725 N of compression produced 0.12 mm of deformation. System compliance was achieved with an in-line compression spring with a stiffness of 23.5 N/mm (Fig. 3). This compliance was necessary to permit the shell to expand when the liner was inserted. Although the exact compliance of the pelvis was not measured, it was noted that, after impaction, the analog pelvis bowed, implying that the strains associated with impaction were dispersed over the entire structure, mitigating the effect of viscoelastic strain relaxation.
The Experiment
Six shells and six liners were used per design to compare deformed (N = 3) and nondeformed (N = 3) shells. Deformation was induced in three shells per design (outside diameter, 52 mm) under 725 N of two-point compression (Fig. 3). Deformation was measured by determining the difference along the load diameter, before and after loading, with use of the dial bore gauge. Corresponding alumina liners (internal diameter, 32 mm) were impacted into the shell with use of three dead drops of a 0.5-kg weight from a distance of 400 mm. The retention strength of the deformed and nondeformed shells was determined according to the method of Tradonsky et al.5 and in accordance with ASTM International (ASTM) Standard F18206. The shells were rim supported and the liners were pushed at 5.1 mm/min with use of a 6.4-mm pin in the apical hole until disassembly occurred, at which time the load was recorded. The taper interface was examined with use of light microscopy for evidence of abrasion or damage.
Source of Funding
CeramTec provided a grant to Orthopaedic Research Laboratories for the design, execution, and presentation of this study, but the authors themselves received no individual funding from CeramTec.
The diametric deformation under 725 N and prior to liner impaction for the shell designs studied and the liner retention strength of the deformed and nondeformed shells are presented in Table I and Figure 4. There was no significant difference (p > 0.05) in the push-out strength between deformed and nondeformed shells with use of a homoscedastic, unpaired, two-tailed Student t test.
Optical analysis of the taper surfaces revealed dramatic differences in taper damage (Figs. 5-A through 5-F). In the two solid alumina liners, metal striping was symmetric and circumferential in the nondeformed shells. This was not true in the deformed shells, where the metal striping was asymmetric and predominantly on the line of loading. In the banded liners, damage to the shell was evident on the line of loading and at the lip in the deformed shells, but no damage was found in the nondeformed shells.
This study suggests that contemporary modular acetabular shell designs that make use of ceramic liners undergo deformation during implantation when the acetabular bed has been underreamed. When ASTM static push-out methods were employed, the liner retention strength of deformed shells did not degrade significantly in comparison with that of nondeformed shells. A larger question, however, is the influence of this deformation on liner retention strength over time when both bone remodeling and cyclic activity occur.
The ideal environment of the laboratory, where surface contaminants, component position, and loading are controlled, is not typical of the operating-room environment, where liner seating is often an issue. Nevertheless, in this optimal situation, shell deformation was consistently observed, as was striping on the outer ceramic liner surfaces. Factors influencing the failure to seat components, as has been reported1, need to be further documented.
This study evaluates both metal-banded and nonbanded ceramic liners of like ceramic composition in which a taper lock is utilized to achieve fixation. A further consideration in this regard may be the mismatch of stiffness between shell and band when both cobalt-chromium-molybdenum and titanium alloy (Ti-6Al-4V) are employed. It is more likely that a cobalt-chromium-molybdenum band will diminish the possibility of any ceramic liner deformation because of its increased stiffness. Any amount of ceramic liner deformation can impact ceramic couple function particular to clearance and sphericity, which in turn may adversely affect fluid-film lubrication and wear. Elimination of a regional clearance around the joint-space opening could account for the observed alterations on the ceramic surface and suggests a potential cause of striping and the squeaking phenomenon. The shell deformation measured in this study could be an initiating factor of these processes. If this is a manifest cause, a logical operative alternative would be the use of stiffer shells or line-to-line reaming and augmentation with fixation screws. 