I. Overview
A. Tribology is the science and technology of friction, lubrication, and wear in interacting surfaces in relative motion.
B. Principles of wear
1. Different modes of wear include adhesive, abrasive, fatigue, and delamination.
2. Adhesive wear predominates in the metal-onpolyethylene bearings used in total hip arthroplasties. Metal-on-polyethylene has been the gold standard in bearing materials for the past 4 decades.
3. Delamination, abrasion, and adhesive wear are commonly seen with total knee arthroplasty.
4. Hip and knee joint simulators are the accepted method of obtaining preclinical test data on wear performance. They have been shown to produce wear particles of similar size and shape to those observed in vivo.
5. Wear is measured gravimetrically on the basis of periodic measurements of the small amount of weight that is lost as material is worn, where 1 million cycles of test equal 1 year of clinical use; however, more recent literature has shown some patients with close to 2 million cycles of activity per year.
*Paul E. Beaule, MD, FRCSC, or the department with which he is affiliated has received miscellaneous nonincome support, commercially derived honoraria, or other nonresearch-related funding from Wright Medical Technology, Inc, and is a consultant for or employee of Wright Medical Technology, Inc.
II. The Hip Joint
A. Wear at the hip joint
1. It is well accepted that polyethylene wear debris produced from the sliding motion of the femoral head within the polyethylene is responsible for osteolysis around total hip implants and subsequent failure of the implant. This has led to development of both hard-on-hard bearings (ceramic-on-ceramic and metal-on-metal) and more wear-resistant polyethylene.
2. In addition to the wear debris generation at the articulating surface, relative motion at modular junctions as well as unintentional contact from impingement can also represent significant sources of wear debris.
B. Biomechanics and kinematics
1. Motion at the hip joint
a. The hip joint is inherently stable because of its relatively rigid ball-and-socket configuration and high conformity. Nonetheless, great mobility is required to perform normal activities of daily living.
b. Hip range of motion takes place in all three planes (sagittal, frontal, and transverse).
i. In the sagittal plane, the flexion-extension motion can exceed 145° with the knee in flexion.
ii. In the frontal plane, the abduction arc can be 120°, and with training it can reach 180°.
iii. In the transverse plane, the arcs of internal and external rotation are approximately 30° and 60°, respectively.
c. Joint angles during various activities have been studied, and high degrees of flexion at the hip joints were measured (squatting: 95.4° of flexion, 28.2° of abduction, 25.7° of external rotation; kneeling: 73.9° of flexion, 25.3° of abduction, 28.1° of external rotation; and sitting
[
Figure 1. Free body diagram of the hip joint. FO = femoral offset, G = gravity force vector, FY Hip = net hip force in the Y direction, FX Hip = net hip force in the X direction, MZ Hip = net hip moment of force about the Z axis, WThigh = total thigh weight vector from the thigh segment center of mass, FY Knee = net knee force in the Y direction, FX Knee = net knee force in the X direction, and MZ Knee = net knee moment of force about the Z axis.]
cross-legged: 85.4° of flexion, 36.5° of abduction, 40.3° of external rotation). The ranges of motion required by most subjects to complete these activities are greater than what can be accommodated by existing prostheses.
2. Surface motion
a. Surface motion in the hip joint can be considered as gliding of the femoral head on the acetabulum.
i. The pivoting of the ball and socket in three planes around the center of rotation in the femoral head produces this gliding of the joint surfaces.
ii. It has been suggested that the gliding mainly occurs at the superior quadrant of the femoral head.
b. If femoral head gliding is incongruent, the shear force may not be parallel to the surface and the articular cartilage may be abnormally compressed, leading to osteoarthritis. This is the case in cam-type femoroacetabular impingement, in which the aspherical femoral head creates outside-in damage of the acetabular cartilage and may account for many cases referred to in the past as idiopathic osteoarthritis.
3. The hip as fulcrum
a. The hip functions effectively as a fulcrum, resulting in a state of equilibrium between body weight and the opposing hip abductors. The length of the lever arm between the femoral head and insertion of the abductors is markedly smaller than that between the femoral head and body weight.
i. The peak hip joint contact force during walking is approximately 2.5 times body weight.
ii. Running creates a joint contact force of approximately 5.2 times body weight during the push-off phase.
iii. Descending stairs is associated with the highest calculated forces and torques across the hip joint, and that may account for its tendency to cause joint symptoms.
b. Gait analyses and free body diagrams (Figure 1) have shown that lack of femoral offset (perpendicular distance from head center to anatomic axis of the femur) significantly increases joint reaction forces.
[
Figure 2. Wear tracks in two different materials after testing on a kneelike wear apparatus for 2 million cycles under identical conditions. A, Conventional polyethylene (1050 resin irradiated at 25 kGy). B,Elevated cross-linked polyethylene (1050 resin irradiated at 65 kGy). The conventional material shows more severe damage. The elastic moduli were 1.0 GPa for the conventional material and 800 MPa for the elevated cross-linked material.]
i. Restoring or increasing femoral offset after hip replacement surgery increases the lever arm of the abductor muscles, thereby reducing the abductor muscle force required for normal gait.
ii. An increase of 10 mm in femoral offset results in a 10% decrease in abductor force requirements and a similar decrease in joint reaction force.
C. Polyethylene in total hip arthroplasty
1. Manufacturing—Ultra-high-molecular-weight polyethylene (UHMWPE) is most frequently manufactured by two methods:
a. Direct molding, in which the polyethylene powder is converted by heat and pressure into a final product
b. Machining from bar stock or sheets of polyethylene
2. Sterilization
a. Historically, most UHMWPE components were sterilized by gamma radiation in air, with dosing varying from 2.5 to 4.0 Mrad. Because of the presence of free radicals, the polyethylene underwent oxidative degradation associated with high wear rates, delamination, and/or gross fracture.
b. Although some polyethylene components were sterilized without radiation, using ethylene oxide or gas plasma, cross-linking is induced by the radiation process; thus polyethylene produced without radiation had inferior wear properties.
3. Highly cross-linked polyethylene
a. Current sterilization techniques are done in inert atmospheres (ethylene oxide or gas plasma) to avoid the introduction of free radicals.
i. The amount of radiation exposure has increased to between 5 and 10 Mrad with either gamma or electron beam radiation. Increasing the level of irradiation has led to greater cross-linking (Figure 2) and therefore improved wear resistance.
ii. Remelting became necessary to eliminate the free radicals created in this process.
b. Remelting versus annealing—It is important to differentiate remelting from annealing, both of which are currently used by different manufacturers.
i. With remelting, the polyethylene is changed from its partial crystalline state to its amorphous state. This can reduce the wear properties of the polyethylene.
ii. With annealing, the polyethylene is heated below the melting point, thus avoiding a reduction in crystallinity but leaving a greater number of free radicals. There are concerns that this process will lead to oxidation of the polyethylene over time.
iii. Current processes sometimes involve general remelting steps with a final ethylene oxide gas sterilizing process.
c. Second-generation highly cross-linked polyethylenes are now available. These polyethylenes use vitamin E, mechanical deformation, or alternating low-radiation doses with annealing to reduce the production of free radicals.
4. Performance
a. In vitro testing of highly cross-linked polyethylene has shown significantly decreased wear rates compared with conventional polyethylene in hip simulator studies.
b. The in vivo performance of cross-linked polyethylene has been very encouraging.
i. Significant wear reduction (ranging from 55% to 95% compared with non-cross-linked polyethylene) has been reported; however, the data are from trials that used a variety of non-cross-linked polyethylenes, some of which had no cross-linking at all and others the standard cross-linking (2.5 to 4 Mrad). This provides some insight into the wide range of relative wear reductions that are reported.
ii. A polyethylene wear rate <0.1 mm per year represents a very low risk of developing osteolysis. Most of the current products on the market have achieved this wear rate.
c. Diminishing material properties have been an undesirable byproduct of the new manufacturing processes. This has been seen with remelting but not annealing.
i. With the remelting process, the final percentage of crystallinity is a bit lower than it was before the polyethylene was remelted. Because crystals arrest crack propagation, there is a corresponding reduction in the fracture toughness/fatigue toughness resistance.
ii. Premature failures have occurred in some rare instances in which the acetabular component was placed vertically and where there is thin polyethylene at the rim. This is especially of concern with the use of larger femoral head sizes in which the polyethylene liner is thinner.
d. Because of the improved wear resistance of these new polyethylenes, larger head sizes (>32 mm) are currently being implanted.
i. In the past, larger head sizes had been associated with increased wear rate due to the increased sliding distance per step, increasing the volumetric wear rate.
ii. At short-term follow-up, the in vivo wear rate of these larger head sizes has improved significantly.
D. Metal-on-metal prostheses
1. Metal-on-metal lubrication theory
a. The general pattern of the wear characteristics for metal-on-metal total hip prostheses is characterized by a run-in wear rate usually occurring in the first million cycles, followed by a steady-state wear rate.
b. The overall volume of wear has been related to the minimum elastohydrodynamic film thickness for mean operating conditions of load, speed, viscosity, and elastic properties of the metals used in each prosthesis, described by an equation called the lambda ratio.
i. The lambda ratio represents the ratio of calculated elastohydrodynamic minimum fluid thickness to composite surface roughness.
ii. If the lambda ratio is ≤1, severe mixed lubrication conditions occur, whereas a lambda ratio ≥3 indicates that fluid film lubrication is contributing substantially to load and hence minimizing friction and wear.
c. Clearance is another factor known to influence wear in metal-on-metal bearings.
i. The clearance is the difference between the diameter of the femoral head and the diameter of the acetabular cup, with diametral clearances <100 μm maximizing the fluid film thickness.
ii. Clearances that are too high lead to increased wear, and clearances that are too low can lead to clamp/equatorial seizing.
d. Increased femoral head diameter increases the entraining velocity, maximizing fluid film lubrication. Even with ideal head diameter and clearance, a mixed film lubrication probably occurs most frequently in vivo.
2. In vivo results
a. Early in vivo data from analysis of retrieved McKee-Farrar prostheses show an average linear wear rate of 0.003 mm/year and 0.004 mm/year for the femoral head and cup, respectively. Interestingly, the larger diameter femoral heads (42 mm versus 35 mm) had a twofold lower mean volumetric wear rate of 0.7 mm3 versus 1.4 mm3 per year.
b. Current-generation metal-on-metal bearings (high-carbon, wrought cobalt-chrome alloy) with a 28-mm-diameter femoral head had an average in vivo wear rate of 27.8 μm/year for the first year of use and after the second year averaged 6.2 μm/year, with heads generally exhibiting a higher wear rate than the cups. There was also a positive correlation between clearance and wear rate.
c. In terms of wear debris, metal-on-metal bearings and corresponding metallic wear debris can release metallic ions in substantially greater concentrations than typically occurs with polyethylene or ceramic, and these ions may form soluble or precipitated organometallic species.
i. Metal-on-metal wear debris particles are on the order of 10 nm to 50 nm.
ii. The corrosion process leads to the production of ions, which are measured in the tissues, blood, and urine.
iii. These ions are being measured prospectively with both resurfacing and stem-type hip devices. These levels are increased in all series.
3. Drawbacks of metal bearings
a. Recently, two types of adverse reactions have been noted in failed metal-on-metal total hip replacements that developed osteolytic responses with persistent pain, the frequency of which is not known.
i. One adverse reaction is a perivascular infiltrate of lymphocytes indicative of a delayed-type hypersensitivity response to the metal wear products and the development of a typical immunologic response.
ii. In addition, other authors have noted the presence of plasma cells, B lymphocytes, and massive fibrin exudation not characteristic of a type-IV delayed-type hypersensitivity reaction and described as an aseptic lymphocyte-dominated vasculitis-associated lesion (ALVAL) or as a lymphocyte-dominated immunological answer (LYDIA).
b. Metal wear particles generated by metal-on-metal hip replacements may have cytotoxic effects through their dispersal throughout the body.
i. The most recent review on cancer incidence with metal-on-metal bearings reported that the only cancer that could potentially occur at an increased rate is leukemia, but this was only during a follow-up period of 5 to 14 years, with the longer follow-up times not showing any increase in the risk of hematopoietic cancers.
ii. More important, the mortality rate in a 20-year follow-up study did not show any difference between metal-on-metal and metal-polyethylene hip replacements.
E. Ceramic bearings
1. Material properties
a. Alumina
i. The alumina used in modern arthroplasty, Al2O3, is a dense, polycrystalline ceramic, obtained from aluminum oxide powder and pressed in a mold at a very high temperature.
ii. It is very stable and chemically inert, unlike zirconia, which needs to be chemically stabilized.
iii. Although alumina is very resistant to compression, it is brittle and is susceptible to fracture.
iv. Recent analysis of clinical data suggests a fracture rate of about 0.012% for alumina heads and inserts.
b. Zirconia
i. To minimize risk of fracture, zirconia was introduced as an alternative.
ii. Zirconia exists in three distinct crystalline phases—monoclinic, tetragonal, and cubic. The phase changes result in a larger variation in volume and significantly decrease the mechanical properties of the material because of the production of cracks. To maintain zirconia in its most stable phase (tetragonal), oxides have been added, creating yttrium-stabilized tetragonal polycrystalline zirconia.
iii. Unfortunately, the clinical wear of zirconia on polyethylene was higher in vivo than what was predicted in vitro, averaging 0.17 mm/year. This is significantly greater than its counterpart alumina/polyethylene, which averages 0.07 mm/year.
2. Wear rates
a. Alumina is a wettable material with clearances of 20 to 50 μm to provide optimal fluid film lubrication.
b. As with metal-on-metal bearings, the alumina/alumina couple exhibits biphasic behavior, with run-in and steady-state wear rates of 1.2 and 0.02 mm3 per million cycles, respectively.
i. Generally, a reduction in grain size and porosity correlates with a lower wear rate.
ii. Of recent concern is the phenomenon of microseparation, leading to stripe wear, probably caused by edge loading during rising from a seated position.
iii. Wear rates in hips with these stripes can be up to 0.3 mm per year or 1.24 mm3 per million cycles.
iv. Recent analysis of alumina wear debris has demonstrated a bimodal distribution, with particle sizes between 5 and 90 nm and between 0.05 to 3.2 μm, with the latter thought to be secondary to microseparation.
c. For the most part, periprosthetic tissues from failed ceramic-on-ceramic prostheses have thinner synovial layers, fewer macrophages, and a lower production of osteolytic substances compared with polyethylene.
3. Drawbacks of ceramic bearings
a. Revision of ceramic prostheses after catastrophic fracture of a ceramic femoral head remains a difficult problem, with some authors reporting survival rates of only 63% at 5 years secondary to aseptic loosening and osteolysis.
b. After catastrophic failure, a complete synovectomy with use of a ceramic or cobalt chrome femoral head and exchange of the acetabular component gave the best results.
c. Dealing with the damaged taper is still controversial. Although some have reported keeping the taper with no recurrence of fracture, if there is any doubt, a new taper is recommended.
III. The Knee Joint
A. Kinematics—Range of motion
1. Knee joint kinematics describe motion taking place in all three planes.
a. In the sagittal plane, the range of motion is the largest, about 160°.
b. Motion in the transverse and frontal planes is linked with the position of the joint in the sagittal plane.
c. The range of rotation increases while the knee flexion increases, reaching a maximum at 90° of flexion.
2. The motion in rotation ranges from 45° to 30° in external and internal rotation, respectively.
3. In the frontal plane, the abduction and adduction motion reaches a maximum of 10° in both directions.
4. The range of motion of the knee in all three planes during walking reaches about 70°, 15°, and 10° in the sagittal, frontal, and transverse planes, respectively.
5. More extreme motion, such as squatting, requires knee flexion up to 160° and external rotation up to 20°.
B. Biomechanics
1. The normal instant center of the knee joint shows a semicircular pathway, which is related to the tibiofemoral surface and ligaments crossing the joint.
a. The surface joint motion occurs between the tibial and femoral condyles and between the femoral condyles and the patella.
b. Studies show that the rupture of the cruciate ligaments or disruption of the tibiofemoral surface, including the menisci, causes a major change in the instant center pathway, leading to articular dysfunction.
2. The slip velocity can quantify the degree of rolling and sliding. During walking, the high slip velocities during heel strike and during swing phase indicate the potential for sliding motion that can produce a greater volume of abrasive wear debris.
3. Successful total knee arthroplasty relies on proper positioning of prosthetic components to restore the mechanical axis of the lower extremity.
4. Gait studies have shown that correcting a varus deformity of the knee to normal alignment after knee arthroplasty could reduce asymmetric loading; consequently, tibial component loosening was also reduced.
C. Polyethylene in knee arthroplasty
1. Wear
a. Polyethylene wear has a multifactoral origin and varied presentation.
b. It can be classified into three types: delamination, abrasive wear, and adhesive wear.
i. Delamination refers to the formation of sub-surface cracks in the polyethylene, which then propagate to the surface over time.
ii. Abrasive and adhesive wear refers to debris formed through micromotion between the polyethylene and the metallic components of the prosthesis. This can occur between the undersurface of the polyethylene liner and the tibial base plate and/or between the polyethylene post and the femoral box in posterior cruciate-substituting knee designs.
c. The generation of polyethylene wear particles initiates an inflammatory process that results in resorption of bone and aseptic loosening. The debris is phagocytized by macrophages from the surrounding tissue, resulting in the release of cytokines that upregulate the production and function of osteoclasts and initiate bone resorption.
d. Research has focused on identifying risk factors for polyethylene wear. The type of polyethylene resin, initial processing, sterilization, and packaging have all been identified as independent factors in wear debris production.
e. Retrieval analysis has been the traditional means of measuring wear rates. Recently, radio-stereometric analysis has been used to quantify wear in vivo; wear rates of 0.13 mm/year have been reported.
2. Sterilization—See section II.C.2, which discusses polyethylene sterilization.
3. Processing
|
a. |
Compression molding involves the administration of heat and pressure to the raw resin to produce the finished articular surface. There is no additional finishing or machining. |
|
b. |
Ram extrusion or machining involves production of sheets of polyethylene that are then shaped or machined to the final product. |
|
c. |
Biomechanical studies have shown that compression molding and ram extrusion produce |
[
Table 1. Comparison of Total Knee Arthroplasty Designs]
|
|
similar wear rates; however, compression molding has a lower susceptibility to fatigue crack formation and propagation. |
|
d. |
In vitro knee simulator studies i. In vitro knee simulator studies demonstrated improved adhesive, abrasive, and delamination resistance with highly cross-linked polyethylene.
ii. These results need to be confirmed in clinical studies of cruciate-retaining and cruciate-substituting designs. |
4. Shelf age and polyethylene thickness
a. Other factors apart from manufacturing affect polyethylene wear. Studies have shown a direct correlation between the age of a polyethylene insert and the generation of wear debris.
i. As previously mentioned, storage of polyethylene in air after gamma radiation allows for the generation of free radicals, which bond to the polyethylene.
ii. Exposure to oxygen on the shelf or after implantation can lead to oxygen free radical bonding to polyethylene, decreasing its mechanical properties.
b. The thickness of the insert implanted also affects the wear rate.
i. Insert thickness <6 mm leads to increased wear rates.
ii. The current recommendation is to use inserts with a minimum polyethylene thickness of 6 to 8 mm.
5. Polyethylene wear and osteolysis
a. Polyethylene wear is the main cause of osteolysis.
b. The emergence of osteolysis as a significant problem in total knee arthroplasty corresponded with the change in knee design from all-polyethylene tibial components to modular compartments that include a metal tibial tray.
i. Current knee systems use a modular tibial component.
ii. Locking mechanisms have been implicated in the production of wear debris as a result of micromotion between the tibial tray and the polyethylene insert.
iii. Backside wear refers to polyethylene debris generation from micromotion between the undersurface of the polyethylene insert and the tibial base plate.
iv. Earlier models of modular tibial components had rough finished tibial trays. This has been shown to increase abrasive wear and debris production.
c. Cam post impingement has been implicated as a source of polyethylene debris in the posterior cruciate-substituting designs. It has also been suggested that rotational forces are transmitted from the cam-post interface to the tibial tray, increasing the backside wear.
6. Design issues (Table 1)
a. Improvements in locking mechanisms between the tibial tray and the polyethylene liner have been made in an attempt to limit backside wear.
b. Contemporary designs also have improved the tibial base plate finish to limit abrasive and adhesive wears.
c. Mobile bearing designs have attempted to decouple the rotation and glide to lower polyethylene stress at each interface.
Top Testing Facts
1. The level of radiation exposure has the greatest influence on the wear properties of polyethylene.
2. Diametral clearance, carbon content, and head size are the main determinants of the wear of metal-on-metal bearings.
3. Alumina is the most stable ceramic bearing surface in vivo, whereas zirconia demonstrates a tendency to switch from its stable tetragonal phase to its monoclinic phase in vivo.
4. Metal particles tend to produce diffuse and perivascular infiltrates of T and B lymphocytes and plasma cells, high endothelial venules, massive fibrin exudation, accumulation of macrophages with droplike inclusions, and infiltrates of eosinophilic granulocytes and necrosis.
5. Backside wear and tibial post impingement are two common sources of wear debris in total knee arthroplasty not usually found in total hip arthroplasty.
Bibliography
Campbell P, Shen FW, McKellop H: Biologic and tribologic considerations of alternative bearing surfaces. Clin Orthop Relat Res 2004;418:98-111.
Collier MB, Engh CA, McAuley JP, Ginn SD, Engh GA: Osteolysis after total knee arthroplasty: Influence of tibial base-plate surface finish and sterilization of polyethylene. Findings at five to ten years postoperatively. J Bone Joint Surg Am 2005;87:2702-2708.
Hemmerich A, Brown H, Smith S, Marthandam SS, Wyss UP: Hip, knee, and ankle kinematics of high range of motion activities of daily living. J Orthop Res 2006;24:770-781.
McKellop H, Shen FW, Lu B, Campbell P, Salovey R: Effect of sterilization method and other modifications on the wear resistance of acetabular cups made of ultra-high molecular weight polyethylene: A hip-simulator study. J Bone Joint Surg Am 2000;82-A:1708-1725.
Ries MD, Pruitt L: Effect of cross-linking on the microstructure and mechanical properties of ultra-high molecular weight polyethylene. Clin Orthop Relat Res 2005;440:149-156.