Medical Devices in Orthopedic Applications
+Author Affiliations
- Address correspondence to: Philip H. Long, Vet Path Services, Inc., 6450 Castle Dr., Mason, OH 45040; e-mail: plong@vetpathservicesinc.com.
Abstract
Orthopedic medical devices have been extremely successful in restoring mobility, reducing pain, and improving the quality of life for millions of individuals each year. Their success is reflected in the worldwide biomaterials market, in which orthopedic devices dominated sales at approximately $14 billion in 2002. Of this, approximately $12 billion was spent on joint replacements. In spite of their overwhelming benefits and successes, orthopedic medical devices are not without risk of adverse effects. Most adverse joint replacement outcomes are thought to be mediated by degradation products generated by wear and electrochemical corrosion. Infection and flaws in device manufacturing are other noteworthy causes of orthopedic device failure. This article illustrates and discusses the uses, general properties, and limitations (including adverse outcomes) of orthopedic biomaterials, which are fundamental to understanding requirements for improving current orthopedic medical devices.
Introduction
Orthopedic Biomaterials Market
Orthopedic biomaterials dominated biomaterial sales at approximately $14 billion in 2002, with an expected growth rate of 7% to 9% annually (Hallab et al., 2004). Most orthopedic medical devices are used in joint replacement or fracture management procedures and this is reflected in their global sales, which totaled approximately $12 billion and $1.5 billion, respectively in 2000 (Hallab et al., 2004). Joint replacement products include devices for hip, knee, ankle, shoulder, elbow, wrist, and finger arthroplasty procedures. Fracture management products include a wide variety of devices including wires, pins, screws, plates, spinal fixation devices, and artificial ligaments. Other orthopedic medical devices include other reconstructive implants, arthroscopy products, electrical stimulation products, and casting products.
Dominant Orthopedic Biomaterials
Despite the large number of existing orthopedic medical devices and applications, there are only a few metals, polymers, and ceramics that dominate the market. Knowing the general properties, uses, and limitations of these materials is important for understanding what is required to improve the performance of implant materials and why only a few dominate the industry. The most common orthopedic biomaterials and their primary uses are listed in Table 1.
Medical Devices in Orthopedic Applications
Table 1Leading orthopedic biomaterials.
Material Primary Uses Metals Ti and Ti alloys Plates, screws, nonbearing surfaces of TJA components Co-Cr-Mo alloys TJA components, including bearing surfaces Stainless steel alloys Wires, pins, screws, plates, cabling Polymers Ultrahigh molecular weight polyethylene (UHMWPE) Low-friction inserts for bearing surfaces in TJA Polymethylmethacrylate (PMMA) Bone cement Ceramics Alumina Bearing-surfaces of TJA components Zirconia Bearing-surfaces of TJA components TJA = total-joint arthroplasty
Metals
Medical Devices in Orthopedic Applications
Table 1Leading orthopedic biomaterials.
Material Primary Uses Metals Ti and Ti alloys Plates, screws, nonbearing surfaces of TJA components Co-Cr-Mo alloys TJA components, including bearing surfaces Stainless steel alloys Wires, pins, screws, plates, cabling Polymers Ultrahigh molecular weight polyethylene (UHMWPE) Low-friction inserts for bearing surfaces in TJA Polymethylmethacrylate (PMMA) Bone cement Ceramics Alumina Bearing-surfaces of TJA components Zirconia Bearing-surfaces of TJA components TJA = total-joint arthroplasty
Table 1
Leading orthopedic biomaterials.
| Material | Primary Uses |
|---|---|
| Metals | |
| Ti and Ti alloys | Plates, screws, nonbearing surfaces of TJA components |
| Co-Cr-Mo alloys | TJA components, including bearing surfaces |
| Stainless steel alloys | Wires, pins, screws, plates, cabling |
| Polymers | |
| Ultrahigh molecular weight polyethylene (UHMWPE) | Low-friction inserts for bearing surfaces in TJA |
| Polymethylmethacrylate (PMMA) | Bone cement |
| Ceramics | |
| Alumina | Bearing-surfaces of TJA components |
| Zirconia | Bearing-surfaces of TJA components |
| TJA = total-joint arthroplasty | |
Certain material properties make metals extremely useful as components in orthopedic medical devices. These properties include corrosion resistance, strength, rigidity, stiffness, long fatigue life, fracture toughness, and biocompatibility. The most important material properties that have lead to widespread use of metal alloys as load-bearing implant materials are strength and corrosion resistance (Hallab et al., 2004). Although Ti alloys, Co-Cr alloys, and stainless steel alloys are commonly used in orthopedic devices, Ti alloys and Co-Cr alloys are the most common metals used in total-joint arthroplasty (TJA) devices.
As with most materials, trade-offs exist with respect to the material properties of various metals and their uses. For example, Co-Cr alloys and ceramics are best suited for bearing surfaces, such as femoral heads, because of their superior hardness and resistance to wear (Hallab et al., 2004). In contrast, the relative softness of Ti alloys compared to Co-Cr alloys results in poor wear and frictional properties; Ti alloys are approximately 15% softer than Co-Cr alloys (Hallab et al., 2004). For this reason, Ti alloys are seldom used where hardness and resistance to wear are most important. Ti alloys are commonly used for non–weight-bearing surface components (femoral necks, stems, and porous coatings) instead of Co-Cr or stainless steel because of their superior resistance to corrosion and because their torsional and axial stiffness are closer to those of bone, resulting in less stress shielding of bone compared to other alloys (Hallab et al., 2004). Greater ductility (threefold better percentage of elongation at fracture) of stainless steel relative to titanium and Co-Cr makes stainless steel ideal for fixation cables used in total-knee arthroplasty procedures (Hallab et al., 2004).
Polymers
Polymers are commonly used in orthopedics as articulating surfaces of joint replacements and as interpositional cementing material between bone and implant surfaces (Hallab et al., 2004). The most common polymers used in TJA products are ultrahigh molecular weight polyethylene (UHMWPE) and polymethylmethacrylate (bone cement or PMMA). UHMWPE possesses excellent sliding properties, high impact strength, high fatigue resistance, and excellent biocompatibility. UHMWPE wears at a rate of approximately 0.1 mm/year, but this is variable and influenced by physical activity of the patient, patient weight, size of femoral head (32 vs. 28 mm), roughness of the metallic counterface, and oxidation of the UHMWPE (Hallab et al., 2004).
Bone cement (PMMA) is used for fixation of joint replacement implants, primarily in older patients (>80 years of age) because the chance of revision is less in this patient population when compared to younger patients (<60 years) and because removing bone cement can be difficult (Hallab et al., 2004). PMMA is often mixed with barium to help visualize the material on radiographs. The barium will not dissolve out during tissue processing, and it is birefringent when viewed with polarized light. PMMA may dissolve in solvents, but if present in histologic sections, it can also be seen with polarized light (Vigorita, 1999). Localized death of bone can occur if bone cement (PMMA) is used to provide mechanical attachment of the device, because a local rise in temperature is created when the monomer cross-links to form the polymer. The heat generated is sufficient to kill bone cells to a depth of approximately 1.0 mm. PMMA must be mixed or prepared at the time of surgery, and it is well known as a sensitizer.
Ceramics
Alumina and zirconia ceramics exhibit superior wear resistance compared to metal–metal or metal–polymer bearing surfaces (Hallab et al., 2004). They are also highly resistant to oxidation and exhibit high stiffness and low frictional properties. Bioactive ceramics such as Bioglass undergo a controlled surface degradation, producing a SiO-rich gel layer in which a Ca-P-rich layer forms (Hench and Andersson, 1993). The Ca-P-rich layer eventually crystallizes as a mixed hydroxy-carbonate apatite that can structurally integrate with collagen.
General Tissue-Implant Responses
All materials elicit some response from the host, so none are truly inert (Hench and Wilson, 1993). The response generally occurs at the tissue-implant interface and depends on many factors. Some important factors include the tissue type, tissue health, tissue age, blood supply, amount of motion, mechanical load, composition of the implant, surface morphology of the implant, and any implant-mediated chemical reactions.
There are four general types of implant-tissue response (Hench and Wilson, 1993; seeTable 2). The first type of response is a toxic response in which the implant material kills cells in the surrounding tissues or releases chemicals that migrate within tissue fluids and cause systemic damage. The second and most common tissue response to an implant is formation of a nonadherent fibrous capsule (Hench and Wilson, 1993). The fibrous tissue forms in an attempt to isolate the implant from the host. This type of response serves to protect the host from an implant and can eventually lead to complete fibrous encapsulation. Orthopedic metal alloys, most polymers, and biologically inactive ceramics such as alumina or zirconia produce this type of interfacial response when implanted into subcutaneous tissue. When implanted into bone, these same materials are usually surrounded by bone and/or fibrous tissue (Figures 1 and 2). The thickness of the fibrous layer depends on a variety of factors. Under ideal conditions, the more chemically inert the material, the thinner the fibrous layer will be; however, it is important to recognize that the thickness of an interfacial fibrous layer is also influenced by motion and fit at the interface (Hench and Wilson, 1993). If a nearly inert implant is loaded such that interfacial movement occurs, the fibrous capsule can become several hundred micrometers thick, and the implant may loosen very quickly (Hench and Wilson, 1993). Loosening generally leads to clinical failure for a variety of reasons, including fracture of the implant, of the bone, or of the bone adjacent to the implant.
Figure 1
Complex polymer implanted into bone. The material exhibits excellent biocompatibility. Bone is in apposition to polymer, and a very thin fibrous layer separates the polymer from the marrow.
Figure 2
Threaded titanium device implanted into bone. The material exhibits excellent biocompatibility. Bone is in apposition to metal (has grown into threads), and a very thin fibrous layer separates the metal from the marrow.
Table 2
Consequences of implant-tissue interactions.
Porous metals, porous ceramics, and hydroxyapatite (HA) coatings on porous metals were developed to prevent loosening of implants (Hench and Wilson, 1993). The growth of bone into the surface pores provides a large surface area between the implant and the surrounding tissue (Figures 3 and 4). This form of attachment is capable of withstanding more complex stresses than nearly inert implants with smooth surfaces. One limitation of porous implants is the need for the pores to be at least 100 micrometers in diameter; this pore size is needed so that capillaries can grow into the pores and provide nutrients to the ingrown tissue (Hench and Wilson, 1993). If movement occurs at the interface of a porous implant, the capillaries can be injured, leading to tissue death, inflammation, and loss of interfacial stability. When the porous implant is a metal alloy, a large interfacial surface area is created, which may provide more area for corrosion of the implant and greater loss of metal ions into the tissues (Hench and Wilson, 1993). The large size and volume fraction of porosity required for stable interfacial bone growth also degrades the strength of the material.
Figure 3
Porous titanium acetabular cup holder. The material exhibits excellent biocompatibility. Bone has grown into pores, resulting in mechanical fixation.
Figure 4
Higher magnification of the porous titanium acetabular cup holder shown inFigure 3, showing bone within pores.
Resorbable implants present a third type of implant-tissue response in that they degrade gradually over time and are replaced by host tissues (Hench and Wilson, 1993). The problem with these materials is meeting the requirements of strength and short-term mechanical performance with an implant while tissue replacement and regeneration are occurring. The implant-resorption rates need to be matched to the tissue-repair rates of the body, which vary greatly depending on a variety of factors (Hench and Wilson, 1993). Some materials may dissolve too rapidly and some too slowly. In addition, large amounts of material must be handled by cells, so the constituents of a resorbable implant must be metabolically acceptable. Examples of resorbable implants include specially formulated polymers composed of poly(lactic acid)-poly(glycolic acid) that are metabolized to carbon dioxide and water (Hench and Wilson, 1993). They function for a time to fill space during wound healing and then dissolve and disappear. Tricalcium phosphate (TCP) ceramics degrade to calcium and phosphate salts and can also be used for space filling of bone (Hench and Wilson, 1993).
The fourth type of interfacial response occurs when a bond forms across the interface between the implant and the tissue; this is termed a “bioactive” interface (Hench and Wilson, 1993). A bioactive implant forms a bond with bone via chemical reactions at their interface. This type of interfacial bond limits or prevents motion between the two materials and can mimic the type of interface that is formed when natural tissues repair themselves (Hench and Wilson, 1993). The bioactive interface requires the implant material to have a controlled rate of chemical reactivity. A key feature of a bioactive interface is that it changes with time, as do normal tissues, which are in a state of dynamic equilibrium.
The bioactive concept has been expanded to include a variety of bioactive materials with a range of bonding rates and thickness of interfacial bonding layers (Hench and Wilson, 1993). They include bioactive glasses such as Bioglass, dense synthetic HA, bioactive composites such as polyethylene-HA, and bioactive coatings such as HA on porous titanium alloy. All of these materials can form an interfacial bond with bone. The time dependence of bonding, the strength of the bond, the mechanism of bonding, the thickness of the bonding zone, and the mechanical strength and fracture toughness differ for the various materials.
A common characteristic of all bioactive implants is the formation of a hydroxyl-carbonate apatite (HCA) layer on their surface when implanted (Hench and Andersson, 1993). The HCA phase is reported to be equivalent in composition and structure to the mineral phase of bone (Hench and Andersson, 1993). The HCA layer grows as polycrystalline agglomerates. Collagen fibrils are incorporated within the agglomerates, binding the inorganic implant surface to the organic constituents. The resulting interface between a bioactive implant and bone is similar to the naturally occurring interfaces between bone and tendons and ligaments.
Implant Loosening
Current total-joint replacement implants last 10 to 20 years (Vigorita, 1999). Revision rates vary, but in general, approximately 10% to 20% of hip arthroplasty operations require revision after 10 years (Vigorita, 1999). Revision is required in large part because of implant loosening. Loosening of an implant can result from a variety of biomechanical factors including excessive activity by the patient, material fatigue, debonding at the tissue-implant interface, stress shielding of bone by the device, inflammation, infection, and osteolysis. The main problem with current total-hip replacement systems is loosening of the acetabular component, which is thought to be primarily caused by inflammation in response to wear debris. Loosening is estimated to occur in as many as 50% of cemented total-hip replacements followed for more than 10 years (Vigorita, 1999). Loosening of cemented implants is usually seen as debonding at the cement–tissue interface, which may be caused by material properties of the cement and/or by inflammation.
Wear Debris
The service life of UHMWPE is limited, and complications occur in about 30% of hip arthroplasties in which UHMWPE is used after 11 years (Gierse and Schramm, 1984). It is estimated that polyethylene wear greater than 0.2 mm annually results in focal osteolysis on radiographs within 5 years (Wan and Dorr, 1996). UHMWPE wear particles vary in size (usually 1 μm in thickness and 4 to 10 μm in length) and are best seen when viewed with transmitted polarized light (Vigorita, 1999). UHMWPE wear particles induce a macrophage, or foreign-body, inflammatory response. Aged UHMWPE may show extensive pitting, cracking, delamination, and other deformities (Figure 5). UHMWPE failure can result in massive, aggressive granulomatous inflammation (Figures 6, 7, and 8).
Figure 5
Aged ultrahigh molecular weight polyethylene (UHMWPE) cup showing cracking and delamination.
Figure 6
Porous titanium acetabular cup holder from an animal with an aged ultrahigh molecular weight polyethylene (UHMWPE) cup. Osteolysis starts along the margins of the joint capsule and migrates along the back of the cup holder.
Figure 7
Periarticular granulomatous inflammation in response to ultrahigh molecular weight polyethylene (UHMWPE) wear debris.
Figure 8
Periarticular granulomatous inflammation shown in Figure 7, viewed with transmitted polarized light showing birefringent particles.
Metal-wear debris particles are generally very small, ranging from approximately 1 to 2 μm in size. The particles may have no effect on surrounding tissue, or they may be associated with inflammation and/or fibrosis. Metal particles may be seen within macrophages or within interstitial connective tissue. Affected tissue may appear black, which is often referred to as metallosis.
Stress Shielding
Bone must be mechanically loaded to remain healthy. Stress shielding occurs when an implant prevents the bone from being properly loaded (Hench and Wilson, 1993). The higher modulus of elasticity of the implant may result in its carrying nearly all the load. Stress shielding weakens bone in the region where the applied load is lowest. Bone that is unloaded will undergo a biological change that leads to bone resorption. The interface between a stress-shielded bone and an implant deteriorates as the bone is weakened. Loosening and/or fracture of the bone, the interface, or the implant may result. The combination of stress shielding, wear debris, and motion at an interface is especially damaging and often leads to failure.
Corrosion
When a metal device is implanted into soft tissue or fluid, a passive metal-oxide film will form on its surface. Depending on the metal, the oxide film can limit oxidation (corrosion) of the underlying metal (Hallab et al., 2004). Iron oxide (rust) that forms on iron is very ineffective at preventing ion exchange (corrosion). In contrast, the Ti oxide film that forms on titanium is extremely effective at preventing ion exchange between the metal and the surrounding soft tissue or fluid. For passive metal-oxide films to be effective at limiting corrosion, they must be compact and fully cover the metal surface, they must have atomic structure that limits migration of ions across the metal-oxide solution interface, and they must remain intact following mechanical stress and abrasion (Hallab et al., 2004). Corrosion, if it occurs, is most likely to occur at sites where the passive oxide film is repeatedly disrupted, such as along the junction and crevices of modular components and where screws contact plates. Continued corrosion at these sites can lead to crack propagation.
Infection
Millions of medical devices are used each year, and in spite of many advances in biomaterials, a proportion of each type of device becomes colonized by bacteria (Costerton et al., 2004). Topical devices (e.g., contact lenses) are colonized by bacteria as soon as they are placed on tissue surfaces; transcutaneous devices (e.g., vascular catheters) are progressively colonized by skin organisms; and even surgically implanted devices regularly become foci of infection (Costerton et al., 2004). It was in the area of topical medical devices that the distinction was first made between colonization, which is the simple presence of microbial films on a surface, and infection that occurs when the presence of a biofilm elicits an inflammatory response (Costerton et al., 2004).
Implanted devices may be colonized by bacteria at the time of surgery or via a hematogenous route from a distant source. The most significant factor in the development of device-related infections appears to be the skill of the surgical team; prosthetic hips become infected in less than 0.2% of cases in large, specialized clinics but in as many as 4% of cases in less proficient facilities (Costerton et al., 2004). Generally, large and complex medical devices that require long and complicated surgery for their placement are at high risk of bacterial infection (Costerton et al., 2004).
Device-related infections may occur almost immediately postsurgery or may be very slow to develop, with overt symptoms occurring months or even years after the device is implanted. Most, if not all, of the characteristics of device-related infections can be explained based on the characteristics of biofilms (Costerton et al., 1999). The biofilm concept was developed in environmental microbiology, and it states that bacteria grow preferentially in matrix-enclosed communities attached to surfaces (Costerton et al., 1978; Costerton et al., 1987). When implanted medical devices become colonized, the microbial biofilms trigger changes in the surrounding tissues, but symptoms are often slow to develop. Symptoms and other indicators of infection, such as white-blood-cell count, may correlate poorly in an implant infection.
The slow development and asymptomatic nature of many device-related infections can be explained by the observation that biofilm bacteria produce few toxins and elicit little inflammatory response (Costerton et al., 2004). In addition, many device-related infections are negative in routine cultures because biofilms release only a limited number of planktonic cells, large biofilm fragments grow up as a single colony on plates, and sessile cells do not grow well on agar surfaces (Costerton et al., 2004). Common bacterial species predominate in device-related infections because they form biofilms very effectively in their natural environments (e.g., skin). The biofilm mode of growth protects the causative agents from both humoral and cell-mediated immunity (Leid et al., 2002), so the organisms are rarely cleared, even by the most active host-defense mechanisms. Exacerbations of device-related infections are caused by the release of planktonic cells. Antibiotics can kill these floating cells and reverse the symptoms of acute infection, but the infection persists because the causative biofilm is resistant to antibiotics (Costerton et al., 2004). Good medical management usually dictates removal of colonized or infected devices (Costerton et al., 2004).
Manufacturing Errors
Several notable flaws in orthopedic-device manufacturing have been reported (Hallab et al., 2004). In 2001, Sulzer Orthopedics, Inc., reported an unintentional mistake in the manufacturing process that affected 31,000 implants. Trace amounts of mineral oil used in the manufacturing process were left on the Inter-Op acetabular shell. The shell, which is covered by a porous titanium coating, subsequently failed to properly bond with the bone into which it was implanted. An estimated 17,500 of the defective parts were implanted in patients nationwide.
Also in 2001, the U.S. Food and Drug Administration (FDA) announced that eight U.S. firms that make hip implants were voluntarily recalling certain hip implants because of a potential fracture problem with a component. The component, a zirconia ceramic femoral head, was recalled by its French manufacturer because it was fracturing at a higher rate than expected in some patients 13 to 27 months after being implanted. The component was the femoral-head portion of the hip prosthesis that connects the femoral stem to the pelvis. The defect was caused by a slight, unintended variation in the manufacturing process when the company began using a new high-throughput, assembly-line–type oven.
Hypersensitivity
Metal hypersensitivity in patients with metal orthopedic implants is reported to be <1% (Hallab et al., 2004). Type IV (cell-mediated) delayed-type hypersensitivity is the response with which orthopedic metal sensitivity is generally associated. Interestingly, dermal hypersensitivity to metal affects 10% to 15% of the population, with nickel being the most common dermal metal sensitizer in humans, followed by cobalt and chromium (Hallab et al., 2004). Although metal hypersensitivity in patients with orthopedic implants is currently viewed as an unusual complication in a few susceptible patients, it is possible that it represents a subtle but common phenomenon that, over time, plays a significant role in implant failure.
Toxicity and Carcinogenicity
Metal toxicity generally applies to soluble forms of the various elements and not to the major chemical species that result from implant degradation (Hallab et al., 2004). Although there are reports of increased metal-ion levels in urine and serum from humans with orthopedic implants, levels have not been associated with metal toxicity. Because tumors associated with medical devices are rare, carcinogenicity tests are only conducted if data from other sources suggest a tendency for tumor development.
No hay comentarios:
Publicar un comentario