John P. Collier

Impact of gamma sterilization on clinical performance of polyethylene in the knee

Damage and rapid wear of the ultrahigh-molecular-weight polyethylene bearings of knee components continue to be major sources of failure of knee prostheses. Despite considerable research into the roles of design, polyethylene thickness and quality, and component alignment, the source of the rapid wear failures has remained a mystery. This study documents elevated oxidation resulting from the use of gamma sterilization in air, the most common sterilization technique used by the orthopaedic implant industry. This oxidation reduces static strength and elongation properties and significantly decreases the resistance of polyethylene bearings to fatigue, a frequent source of early damage of many of these devices.

Analysis of the failure of 122 polyethylene inserts from uncemented tibial knee components.

The extent of wear of retrieved, polyethylene tibial components appears to be related to design. To test this observation, 122 tibial inserts were graded for wear, and new components of several designs were tested for contact stress using Fuji film. Finite element analysis provided insight into subsurface stresses. Significant wear was seen in 61.5% of the tibial inserts examined. The presence of unconsolidated polymer powder was seen in 44% of the tibial inserts and was found to be statistically correlated with severe wear of the articulating surface. Contact stress in several noncongruent designs was found to exceed the yield strength of polyethylene. There was a positive correlation between the intensity of wear and the level of contact stress, with noncongruent designs having greater wear than fully congruent geometries. In the non-congruent designs, the thinner polyethylene components showed greater wear than thicker polyethylene inserts of the same design.

Corrosion between the components of modular femoral hip prostheses

We studied the tapered interface between the head and the neck of 139 modular femoral components of hip prostheses which had been removed for a variety of reasons. In 91 the same alloy had been used for the head and the stem; none of them showed evidence of corrosion. In contrast, there was definite corrosion in 25 of the 48 prostheses in which the stem was of titanium alloy and the head of cobalt-chrome. This corrosion was time-dependent: no specimens were corroded after less than nine months in the body, but all which had been in place for more than 40 months were damaged. We discuss the factors which may influence the rate of these changes and present evidence that they were due to galvanically-accelerated crevice corrosion, which was undetected in previous laboratory testing of this type of prosthesis.

Tibial insert undersurface as a contributing source of polyethylene wear debris.

Sixty-seven ultrahigh molecular weight polyethylene tibial inserts from cementless total knee arthroplasties were retrieved at autopsy and revision surgery and analyzed for evidence of articular and nonarticular surface wear after a mean implantation time of 62.8 months (range, 4-131 months). Polyethylene cold flow and abrasive wear on the nonarticular insert surface (undersurface) were assigned a wear severity score (Grade 0-4). The severity of articular wear was assessed quantitatively and graded. Corresponding prerevision radiographs were evaluated for evidence of tibial metaphyseal osteolysis and osteolysis around tibial fixation screws. Exact nonparametric conditional inference methods were used to establish correlations between different variables and the occurrence of tibial metaphyseal osteolysis. Severe Grade 4 wear of the tibial insert undersurface was associated with tibial metaphyseal osteolysis or osteolysis around fixation screws. Time in situ statistically was related to Grade 4 undersurface wear and tibial metaphyseal osteolysis. The occurrence of tibial osteolysis was not related statistically to articular wear severity, insert thickness, or implant type. The main articulation between the femoral implant and ultrahigh molecular weight polyethylene insert has been assumed to be the primary source of polyethylene debris contributing to osteolysis and total knee arthroplasty implant failure. The undersurface of the insert is an additional source of polyethylene debris contributing to tibial metaphyseal osteolysis. To lessen polyethylene debris produced at this modular interface, the tibial implant locking mechanism should fix the insert firmly to the metal backing to decrease relative micromotion. Because motion between the insert and metal backing may be inevitable, the wear characteristics of the inner tray surface should be optimized to minimize wear debris production at this other articulation.

Corrosion at the interface of cobalt-alloy heads on titanium-alloy stems.

The combination of a cobalt-alloy head on a titanium-alloy femoral hip stem is widely accepted for press-fit and biologic fixation applications. Examination of 30 components retrieved at periods of 0.5 to 66.9 months for histologic examination of tissue ingrowth revealed that 56.6% of the tapered connections between head and stem showed evidence of crevice corrosion leading to concerns of metal ion release and the potential failure of head to stem fixation.

Comparison of cross-linked polyethylene materials for orthopaedic applications.

Cross-linked polyethylenes are being marketed by orthopaedic manufacturers to address the problem of osteolysis caused by polyethylene particulate wear debris. Wear testing of these cross-linked polyethylenes in hip simulators has shown dramatic reduction in wear rate compared with standard ultrahigh molecular weight polyethylene, either gamma irradiated in air or nitrogen - or ethylene oxide-sterilized. However, this reduction in wear rate is not without cost. The cross-linking processes can result in materials with lower mechanical properties than standard ultrahigh molecular weight polyethylene. To evaluate the effect of the various cross-linking processes on physical and mechanical properties of ultrahigh molecular weight polyethylene, commercially available cross-linked polyethylenes from six orthopaedic manufacturers were tested. This study was the culmination of collaboration with these manufacturers, who provided cross-linked polyethylene for this study, wear characteristics of the material they provided, and review of the physical and mechanical properties measure for their polyethylene. Cross-linked materials were evaluated as received and after an accelerated aging protocol. Free radical identity and concentration, oxidation, crystallinity, melt temperature, ultimate tensile strength, elongation at break, tensile stress at yield, and toughness are reported for each material. By comparing these physical and mechanical properties, surgeons can evaluate the trade-off that results from developing materials with substantially lower wear rates.

Overview of polyethylene as a bearing material: comparison of sterilization methods.

Polyethylene has been used for more than 30 years as an orthopaedic bearing material; however, there has been recent concern regarding the early failure of a small percentage of the polyethylene bearings. The damage seen in some retrieved polyethylene components has been linked to gamma radiation sterilization in air, which was widely used by the industry for years. Gamma radiation in air has been documented to cause an increase in oxidation and degradation of mechanical properties with time. The degradation of polyethylene initiated by gamma sterilization in air has led the orthopaedic industry toward alternative sterilization methods, including gamma radiation in an inert gas or vacuum environment, ethylene oxide gas sterilization, and gas plasma sterilization. For many of these alternative techniques, little clinical performance data exist. This study is a comparative evaluation of sterilization methods using the same analytic techniques that have been used to document the effects of gamma sterilization in air on polyethylene. Fourier transform infrared spectroscopy, electron spin resonance, and uniaxial tensile testing are used to compare, respectively, the oxidation levels, free radical concentration, and mechanical properties of material sterilized by each method. The polyethylene is evaluated before sterilization, poststerilization, and postartificial aging. All examined alternative sterilization methods, when compared with gamma sterilization in air, caused less material degradation during a components preimplantation shelf life.

Macroscopic and microscopic evidence of prosthetic fixation with porous-coated materials.

The host response to porous-coated prostheses appears favorable; there is little evidence of any adverse tissue response or significant osteoclastic activity except in grossly loose specimens. While the nature of retrieval specimens makes any statistical correlation problematic, some generalizations can be made. Femoral hip prostheses are most likely to present bone ingrowth along the lateral quadrant of their porous coating. The frequency of bone ingrowth of femoral components was nearly twice that of acetabular devices. Pore size, geometry, and porous-coating composition did not appear to influence the appearance of bone and fibrous tissue ingrowth. Direct bonding of bone to the uncoated portion of the prosthesis was rarely seen and occurred only in closest proximity to the porous-coated regions. Indications of pain and looseness are evidence that fibrous tissue ingrowth alone is not always sufficient to ensure stability. Additionally, some bone-ingrown prostheses were retrieved because of pain, which leads to the conclusion that local bone ingrowth cannot ensure a general freedom from pain, especially with partially coated prostheses. Bone and fibrous tissue response to the porous coatings generally consists of interdigitation, while the response to uncoated regions is fibrous tissue encapsulation. Burnishing the distal tips of many of the partially coated femoral prostheses is an indication of relative motion in that region, which may be a potential source of pain.

Mechanisms of failure of modular prostheses.

The expectations of wear and longevity of total hip components are based in large part on Charnleys early work. The evolution of the total hip from the one-piece, all-polyethylene acetabular component and fixed-head femoral component to the myriad of parts that comprise many of todays total hip designs has brought with it an array of potential mechanisms for failure that were not present in the earlier design. The risk/benefit ratio of these new designs may need to be reevaluated based on the additional mechanisms for failure that they provide. One hundred eleven acetabular hip prostheses and 139 femoral prostheses, all of modular configuration, retrieved by surgeons in the field, and sent for histologic examination, were analyzed for this study. A number of component characteristics were found to be correlated to early failure. These included acetabular designs with thin polyethylene bearings, poor fixation of the polyethylene to the metal shell, and geometries that permitted a moment to be applied to the bearing insert, tending to cause it to rotate in the metal shell. Modular femoral components were observed to be susceptible to corrosion, with titanium-alloy stems mated to cast cobalt-alloy heads at greatest risk attributable to a galvanic effect. All modular connections of femoral and acetabular components are at risk for disassociation and fretting; therefore, clever design and precision machining are necessary to produce prostheses in which the benefits of modularity exceed the risks.

Evaluation of Oxidation and Fatigue Damage of Retrieved Crossfire Polyethylene Acetabular Cups

BACKGROUND Crossfire cross-linked polyethylene is produced differently from other cross-linked polyethylene materials; a below-melt-temperature annealing process is used with the goal of avoiding compromised mechanical properties. The present study was performed to evaluate retrieved Crossfire acetabular cups to determine whether they had oxidized and to what extent oxidation might have influenced their clinical performance. METHODS Eleven acetabular cups were received at retrieval and a twelfth acetabular cup was received two years post-retrieval over a period of four years. None were retrieved because of polyethylene wear or fatigue. The cups had been in vivo from 0.1 to 5.3 years. Each was examined visually, clinical fatigue damage was rated, and oxidation was measured with use of Fourier transform infrared spectroscopy. RESULTS The cups exhibited oxidation that varied with its location on the cup: the oxidation value was generally low on the articular surface but more than an order of magnitude higher value on the rim. Maximum rim oxidation correlated significantly with the time in vivo (Spearman rho = 0.734, p = 0.010). Oxidation was identified visually by a white band in thin sections on the rim of seven of the cups and on the articular surface of one of these seven cups. Six of the seven cups also exhibited clinical fatigue damage. Eight of the twelve cups exhibited evidence of impingement or dislocation. CONCLUSIONS Acetabular cups made of Crossfire polyethylene oxidized to a measurable degree. The oxidation-related reduction of polyethylene mechanical properties was sufficient to allow the fatigue damage seen in these retrieved cups.