Orthopaedic Knowledge Home Study by R. Alexander

Orthopaedic Knowledge Update 8 Chapter 4 Musculoskeletal Biomechanics It has been established that mesenchymal progenitor Efunda: Engineering Fundamental Website. Available cells (MPCs) from both humans and animals can differ- at: http://www.efunda.com. Accessed June, 2004. entiate into a variety of cell types, including fibroblasts, osteoblasts, and endothelial cells. These cells are respon- This website provides information that is helpful in the un- sible for tissue wound healing. Bone marrow, blood, derstanding of biomechanics. Information includes formulas muscle, or adipose tissue are potential sources of MPCs for beam deflection, bending, and twisting and on stress and for cell therapy. For cell therapy, MPCs are isolated strain parameters for materials. from the bone marrow, cultured, and finally trans- planted into host tissues. MPCs appear to retain their Keaveny TM, Yeh OC: Architecture and trabecular developmental potential even after extensive subcultur- bone: Toward an improved understanding of the biome- ing. Implantation of MPCs into an injured tendon signif- chanical effects of age, sex and osteoporosis. J Musculo- icantly improves its mechanical properties. skeletal Neuronal Interactions 2002;2:205-208. The basis for future treatments eventually may use a This study reviews what is known about architectural combination of approaches to treat injuries that include changes in trabecular bone associated with age, gender, and os- seeding cells on a scaffold that is conditioned with the teoporosis and the role of these changes in the mechanical right combination of mechanical stimuli, growth factors, properties of tissue. Recent developments in three-dimensional and gene therapy. Because these experiments will re- high-resolution imaging technologies have provided more accu- quire a combination of techniques that simultaneously rate measures of quantitative metrics of architecture, thereby address the structure, biochemistry, and biology of a tis- providing new data and raising questions about earlier conclu- sue to enhance healing, a significant effort must be sions. made to encourage collaboration among investigators from different disciplines. By combining appropriate en- NSBRI: National Space Biomedical Research Institute gineering mechanics with other basic sciences, improved Website. Available at: http://www.nsbri.org. Accessed outcomes can be achieved. June, 2004. Annotated Bibliography This website provides a description of areas of research such as bone loss, muscle atrophy, and other current research projects. An KN: In vivo force and strain of tendon, ligament and Panjabi MM, White AA III: Biomechanics of the Muscu- capsule, in Guilak F, Butler DL, Goldstein SA, Mooney loskeletal System. New York, NY, Churchill Livingstone, DJ (eds): Functional Tissue Engineering. New York, NY, 2001. Springer, 2003, pp 96-105. This book describes the principles of biomechanics in a In vivo force and strain measurements in soft tissue were simple manner and is suitable for orthopaedic residents and presented in six groups: tendon tension, ligament deformation, fellows. capsular pressure, tendon surface friction, soft-tissue stress, and soft-tissue strain. Puska MA, Kokkari AK, Närhi TO, Vallittu PK: Me- chanical properties of oligomer-modified acrylic bone Ateshian GA, Hung CT: Functional properties of native cement. Biomaterials 2003;24:417-425. articular cartilage, in Guilak F, Butler DL, Goldstein SA, Mooney DJ (eds): Functional Tissue Engineering. Mechanical properties of oligomer-modified acrylic bone New York, NY, Springer, 2003, pp 46-68. cement with glass fibers were studied. The three-point bending test was used to measure the flexural strength and modulus of Important mechanical properties of the cartilage including the acrylic bone cement composites using analysis of variance viscoelasticity, anisotropy, tension-compression nonlinearity, between groups. A scanning electron microscope was used to and inhomogeneity were assessed. Valuable measures of the examine the surface structure of the acrylic bone cement com- functionality of the tissue-engineered constructs were in- posites. cluded. Butler DL, Dressler M, Awad H: Functional tissue engi- Woo SL, Abramowitch SD, Loh JC, Musahl V, Wang JH: neering: Assessment of function in tendon and ligament, Ligament healing: Present status and the future of func- in Guilak F, Butler DL, Goldstein SA, Mooney DJ tional tissue engineering, in Guilak F, Butler DL, Gold- (eds): Functional Tissue Engineering. New York, NY, stein SA, Mooney DJ (eds): Functional Tissue Engineer- Springer, 2003, pp 213-226. ing. New York, NY, Springer, 2003, pp 17-34. This chapter discusses the issues that confront researchers Properties of normal ligaments, including their anatomic, in fabricating structures from cells. State-of-the-art applica- biologic, biochemical, and mechanical properties, and changes tions of biologic-microelectromechanical systems, imaging, and that occur after injury, were described. Functional tissue engi- other technologies are described. neering methods and preliminary findings were presented. American Academy of Orthopaedic Surgeons 55

Musculoskeletal Biomechanics Orthopaedic Knowledge Update 8 Classic Bibliography Martin BR, Burr DC, Sharkey NA: Skeletal Tissue Me- chanics. New York, NY, Springer, 1998. Burstein AH, Reilly DT, Martens M: Aging of bone tis- sue: Mechanical properties. J Bone Joint Surg Am 1976; Mow VC, Ateshian GA: Lubrication and wear of diar- 58:82-86. throdial joints, in Mow VC, Hayes WC (eds): Basic Or- thopaedic Biomechanics, ed 2. New York, NY, Raven Ding M, Dalstra M: Danielsen CC, Kabel J, Hvid J, Press, 1997, pp 275-316. Linde F: Age variations in the properties of human tib- ial trabecular bone. J Bone Joint Surg Br 1997;79:995- Ozkaya N, Nordin M: Fundamentals of Biomechanics, ed 1002. 2. New York, NY, Springer-Verlag, 1998. Hall SJ: Basic Biomechanics, ed 3. New York, NY, McGrawHill, 1999. 56 American Academy of Orthopaedic Surgeons

5Chapter Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Timothy M. Wright, PhD Introduction in vivo. For many other applications, however, the situa- tion is not as straightforward. Synthetic materials play a prominent role in ortho- paedic surgery because of the continuing need to re- The goal of improving implant wear resistance is a place, stabilize, or augment damaged musculoskeletal difficult task. The introduction of new forms of existing tissues. Materials used for orthopaedic devices must be materials or alternative bearing materials is hampered safe and effective; therefore they must be biocompati- by the lack of strong scientifically based relationships ble, resistant to corrosion and degradation, possess su- between specific material properties (measured using perior mechanical and wear properties, and meet high standardized laboratory tests) and in vivo wear behav- quality standards—all at a reasonable cost. The inter- ior. Knowledge has been gained by analyzing the play between synthetic materials and the surrounding stresses that occur in bearing materials under realistic environment is an important factor to consider when us- geometries and loading conditions; however, the link be- ing such materials to temporarily stabilize a fracture or tween specific wear mechanisms and the controlling ma- to permanently replace a structure such as the hip joint. terial properties remains circumstantial. Thus, material selection cannot be made on the simple basis of specifi- Implant wear is the major complication that limits cations such as elastic modulus, fracture toughness, or the longevity of total joint arthroplasties. Substantial yield stress. Similarly, standard laboratory experiments clinical evidence exists that links the release of large on simple geometries, such as pin-on-disk tests, do not amounts of submicron particulate debris from articular adequately recreate the mechanical stresses that the ma- and modular interfaces in implant components to subse- terial will experience in vivo. quent osteolysis and implant loosening. Bioengineering solutions to the wear problem are based on two ap- Because of these limitations, screening tests using proaches: (1) replacing the conventional metal-on-ultra- hip and knee joint simulators have become the accepted high molecular weight polyethylene (UHMWPE) artic- approach for providing preclinical test data on wear per- ulation with alternative combinations of bearing formance. Wear is measured gravimetrically on the basis materials with improved wear resistance, and (2) using of periodic measurements of the small amount of weight improved designs aimed at lowering contact stresses and that is lost as material is worn from the articular surface sliding distances between moving parts. Only long-term during the simulator test. Joint simulators are validated clinical experience will ultimately establish the efficacy in the sense that, under certain kinematics and loading of these approaches; however, laboratory results and conditions, they produce worn polyethylene surfaces short-term clinical experience suggest that such ap- and generate wear particle sizes and shapes similar to proaches are beneficial in reducing implant wear. those observed on retrieved implants. Hip simulators also produce wear rates over the course of the test that Appreciation of the basic science behind these and generally agree with clinical wear rates determined from other uses of biomaterials in orthopaedic surgery re- component thickness changes observed on serial radio- quires knowledge of the basic structure and composition graphs (assuming one million cycles of test equals 1 year of these materials and an understanding of how a mate- of clinical use). rial’s structure and composition determine its ability to meet necessary performance criteria essential to its clin- The reliance on joint simulator tests is especially im- ical efficacy. For many applications of biomaterials in or- portant because the clinical ramifications of improved thopaedics, the failure criteria for the material can be wear are not known until many years after the introduc- measured and directly compared with the expected me- tion of a new material. If a bearing material is intended chanical burden the material will be subjected to for use to produce a meaningful reduction in wear, its effective- ness can only be shown through long-term studies; how- ever, this requirement makes it difficult for both indus- American Academy of Orthopaedic Surgeons 57

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Orthopaedic Knowledge Update 8 Table 1 | Benefits and Disadvantages of Bearing Material Combinations Bearing Combination Benefits Disadvantages Metal-on-polyethylene Long-term clinical results Osteolysis Design effects well known May be unsuitable for young patients Ethylene oxide or gas plasma sterilized Long-term clinical results Limited shelf life before implantation polyethylene-on-metal or ceramic Low wear Wear rates not as low as elevated cross-linked Excellent resistance to degradation Polyethylene sterilized in low oxygen/inert polyethylene environment Excellent resistance to degradation Wear rates not as low as elevated cross-linked Excellent resistance to degradation polyethylene Low wear Wear rates not as low as elevated cross-linked Elevated cross-linked polyethylene-on-metal or Excellent wear resistance polyethylene ceramic Excellent resistance to degradation Possible residual free radicals Metal-on-metal Usually low wear Short-term clinical use Possible lubrication film Reduced toughness could be problematic High cost Ceramic-on-polyethylene Low wear Alumina-on-alumina Good resistance to third body wear Sometimes high wear Local and systemic accumulation of metallic Usually low wear Excellent biocompatibility debris and ions Possible lubrication film Component fracture/difficult revision High cost Sometimes high wear Component fracture High cost (Adapted from McKellop HA: Bearing surfaces in total hip replacements: State of the art and future developments. Instr Course Lect 2001;50:174.) try and the Food and Drug Administration (FDA) to ide, which serves as a protective passive layer between bring improved technologies to the marketplace safely the corrosive body environment and the bulk steel. and expeditiously. Current evaluation of the perfor- mance of new bearing materials must be centered on Cobalt alloys are composed primarily of cobalt, their intended benefits and disadvantages (Table 1) and chromium, and molybdenum. Chromium is added for in- the results of laboratory and short-term clinical tests. creased hardness and corrosion resistance. As is the case with stainless steel, the chromium forms a strongly ad- Metallic Materials herent passive oxide film. Molybdenum is added to pro- vide high strength. Standard cobalt alloy contains signif- Stainless steels, cobalt alloys, and titanium alloys have icant amounts of carbon that allows for the formation of been used in orthopaedic devices for decades. They are hard carbides, which strengthen the alloy and improve generally corrosion resistant and biocompatible and its wear resistance. The addition of nickel forms an alloy possess mechanical properties sufficient for use as struc- suitable for forging, which further enhances the materi- tural load-bearing implants. These materials are fabri- al’s strength. cated using a wide variety of techniques (including cast- ing, forging, extrusion, and hot isostatic pressing) which The most commonly used titanium alloy for ortho- lend flexibility in terms of both mechanical properties paedic applications is titanium-aluminum-vanadium al- and shape. loy, often referred to as Ti-6Al-4V because the primary alloying elements, aluminum and vanadium, comprise Stainless steels are predominantly iron-carbon al- about 6% and 4%, respectively, of the alloy. Titanium loys. Carbon is added to allow the formation of metallic has the ability to self passivate, forming its own oxide carbides within the microstructure. Carbides are much that imparts a high degree of corrosion resistance. Oxy- harder than the surrounding material, and a uniform gen concentration is kept very low to maximize strength distribution of carbides provides strength. Additions of and ductility. Despite long-term clinical evidence of ex- other alloying elements, such as molybdenum, stabilize cellent biocompatibility, concern that the release of va- the carbides. Chromium provides the stainless quality to nadium (a cytotoxic element) could cause local and sys- stainless steel. It forms a strongly adherent surface ox- temic complications has led to the introduction of other 58 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty titanium alloys in which vanadium is replaced by more joints proved superior and as carcinogenic concerns inert elements such as niobium. over local and systemic accumulation of particulate and ionic metallic debris grew. These early failures were Stainless steel is the material that is most susceptible partly the result of poor metallurgy and less than opti- to both galvanic and crevice corrosion. The area be- mum implant design. Casting of metallic alloys often re- tween the screw heads and countersunk holes in stain- sulted in microstructures with large grains and poorly less steel bone plates is a common location for crevice distributed carbides, resulting in inhomogeneity in sur- corrosion. Continual removal of the passive layer can be face hardness. Early metal-on-metal designs often had caused by fretting and can lead to the corrosion of all small head to neck ratios; therefore, impingement be- three metallic alloys. Modular metallic devices, including tween the neck of the femoral component and the rim femoral components for total hip replacement and of the acetabular components was a common occur- stemmed components for revision knee replacement, rence. provide both the micromotion that causes fretting and the enclosed environment needed for crevice corrosion. More recently, improved metallurgy and manufactur- Damage caused by corrosion and fretting often is found ing techniques have led to the reintroduction of metal- on retrieved modular components, and is associated on-metal bearings for hip replacement. Cobalt- with local and systemic accumulations of alloy elements chromium alloys with well-controlled grain sizes and such as titanium and chromium. finely distributed carbides provide superior hardness and wear resistance compared with earlier versions of Corrosion and strength concerns have largely rele- the alloy and to implants made of stainless steel and ti- gated the use of stainless steel components to tempo- tanium alloy. Laboratory evidence from hip joint simu- rary implant devices, such as fracture plates, screws, and lator studies has confirmed that these improved bearing hip nails; however, with appropriate designs, stainless surfaces can provide low friction and low wear articula- steel is still used in permanent implants such as Charn- tions. Clearance and conformity between the mating ley style femoral components for hip replacement. Co- surfaces are now recognized as important parameters balt alloys, because of their high strength and their abil- that must be controlled as part of the design and manu- ity to be polished to a very smooth surface finish, are facturing processes. the most common materials used for metallic total joint arthroplasty components. The strength of cobalt-chromium alloys (compared with polyethylene) and their increased toughness over Titanium alloys possess roughly one half of the elas- ceramics provides additional benefits from the stand- tic modulus of stainless steel or cobalt alloy materials; point of hip implant design. For example, the wall thick- therefore, for the same component design, a titanium ness of solid metallic acetabular components can be version will have one half the structural stiffness smaller than modular polyethylene and metal or ce- (whether subjected to axial, bending, or torsion loads). ramic and metal implants, so larger head sizes can be in- This property makes titanium alloys attractive in situa- corporated, providing an advantage for patients where tions in which more load is to be shared with the sur- joint stability is a concern. Similarly, the ability to manu- rounding bone. The low modulus and excellent strength facture large metallic shells allows for surface replace- of these alloys make them useful in a host of devices for ment of the hip joint, which is a bone-conserving opera- trauma, spinal fixation, and arthroplasty. The latter tion with particular indications for young, active patients application, however, no longer extends to articular sur- with good bone stock in the femoral head and neck. faces because significant scratching and wear of tita- nium alloy femoral heads has been observed, particu- Laboratory and analytical evidence suggest that larly in the presence of third body wear. wear rates of metal-on-metal bearings can be markedly affected by head size. Unlike metal-on-polyethylene Metal-on-Metal Bearings bearings, in which larger head sizes tend to increase wear because of increased sliding distances between the The problems of wear and osteolysis in total joint ar- articulating surfaces, wear decreases significantly in throplasties have led to a resurgence of interest in metal-on-metal bearings with larger heads. Large femo- metal-on-metal bearing surfaces. Metal-on-metal articu- ral head-acetabular component combinations seem to lations were among the first to be used in total hip ar- form a protective lubricating film that separates the sur- throplasty and had clinical success in the 1960s and faces. Smaller head sizes (≤ 28 mm) have much less or 1970s in designs such as the McKee-Farrar (Howmedica, no separation and therefore increase the propensity for Limerick, Ireland) hip replacement. Recent studies sug- adhesive and abrasive wear. gest that many of these early implants have performed well, even after implantation intervals approaching 30 Corroborating clinical evidence of the effect of head years. size on wear rates in metal-on-metal hip replacements does not yet exist. With most modern versions of alter- Metal-on-metal hip joint replacements fell from fa- native bearings, clinical results are available only for vor when the clinical results of polyethylene-on-metal short- to intermediate-term usage. To date, few serious American Academy of Orthopaedic Surgeons 59

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Orthopaedic Knowledge Update 8 wear-related problems such as osteolysis have been of eliminating degradation. These techniques do not found. Nonetheless, serum concentrations of metallic el- have the same beneficial impact on wear. Sterilization ements including chromium, cobalt, and molybdenum techniques that eliminate irradiation, such as exposure are found in increased levels in patients receiving metal- to gas plasma and ethylene oxide, also eliminate the po- on-metal hip replacements compared with control tential benefit of the additional cross-linking produced groups and patients receiving metal-on-polyethylene by irradiation. Hip joint simulator studies have shown a joint implants. The same cytokines found in abundance higher rate of wear for components sterilized with gas around failed metal-on-polyethylene joints have been plasma and ethylene oxide compared with gamma irra- identified around failed metal-on-metal joints, suggest- diation; however, the long-term clinical relevance of ing that the same biologic pathways for osteolysis exist these differences remains unknown. Little is known in these joints. Laboratory evidence also suggests that about the impact of irradiation-free sterilization tech- increased adhesion can occur between metal-on-metal niques on the pitting and delamination surface damage bearing surfaces at the startup of motion after a resting that occurs in total knee polyethylene components. period (common in patients’ daily activities); the clinical Studies of retrieved tibial components originally steril- ramifications remain unknown. ized in ethylene oxide have shown less evidence of cracking than conventionally sterilized components. Only longer and more extensive clinical experience will determine if the reduced volumetric wear with metal- Other sterilization techniques that use irradiation in on-metal bearings will lower the incidence of osteolysis an inert or low-oxygen environment are more beneficial and wear-related failures. Data will soon become available in improving wear resistance, based on improved wear because metal-on-metal hip joints have received approval rates found in hip simulator studies. The improvement by the FDA for commercial distribution in the United in wear behavior is probably the result of the cross- States. Surface replacements using metal-on-metal bear- linking of polyethylene chains that occurs with irradia- ings are currently in the investigational stage.An early re- tion. However, free radical production is still a detri- port (2- to 6-year follow-up) showed a revision rate of 3% mental result, and postirradiation thermal treatment is because of loosening and femoral neck fracture. performed to quench free radicals and prevent degrada- tion that could occur with subsequent exposure to an Metal-on-metal bearings have not been used for oxygen environment. other joints, such as the knee. Most other joints require very different design concepts and are subject to addi- The fabrication technique used in manufacturing tional wear mechanisms for which metal-on-metal sur- polyethylene total joint components also affects wear faces have few advantages. behavior. Direct compression molding in which polyeth- ylene powder is heated and pressed in molds to form a Ultra-High Molecular Weight Polyethylene finished product results in polyethylene components re- sistant to oxidative degradation. In clinical use for more Clinical follow-up for polyethylene-on-metal bearings than 20 years, directly molded hip and knee replace- now extends beyond 25 years, which is longer than for ment components have shown favorable wear resis- some other bearing types. Considerable biologic and tance, based on clinical evidence, studies of wear dam- clinical evidence shows that polyethylene wear debris age in retrieved components, and laboratory wear tests elicits a deleterious biologic response culminating in os- compared with components machined from extruded teolysis. Biomaterial approaches to combat polyethylene polyethylene stock. The lower elastic modulus that re- wear that do not require the introduction of alternative sults from compression molding may be a key factor. bearing materials are attractive because they can capi- Under the same load conditions, a lower modulus poly- talize on the relatively low cost of manufacturing poly- ethylene will have larger contact areas and will experi- ethylene components, the material biocompatibility and ence lower stresses than a higher modulus polyethylene toughness in bulk form, and an easier path to regulatory under the same load conditions. approval. In the past, polyethylene components have been sterilized by exposure to gamma radiation at a The most significant recent alteration in fabrication higher dose than 25 kGy in an ambient environment. It techniques for manufacturing polyethylene joint arthro- is now known that this technique causes oxidative deg- plasty components is the inclusion of elevated levels of radation. The exposure to radiation energy causes chain radiation to induce even higher levels of cross-linking scission, chain cross-linking, and the creation of free than occur with the conventional sterilization dose radicals in the material, resulting in a decrease in molec- (25 kGy). Advantages of elevated cross-linking include ular weight, an increase in density, and detrimental al- significantly reduced wear (as shown in hip and knee terations in mechanical properties that add significantly joint simulator test) and, in the case of knee joints, a re- to the wear problem. sistance to pitting and delamination. Wear rates of es- sentially zero have been shown, for example, in hip sim- In response to these findings, several alternative ster- ulator tests for elevated cross-linked polyethylenes ilization techniques have been introduced with the aim (Figure 1). Few well-controlled clinical studies in the 60 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty peer-reviewed literature exist to date on the in vivo Figure 1 The wear rates of acetabular cups tested in a hip simulator decrease with wear behavior of these materials. Short-term results sug- exposure to higher doses of radiation applied to the components prior to testing. gest an improvement in wear; however, a longer (Adapted with permission from McKellop H, Shen FW, Lu B, Campbell P, Salovey R: follow-up period is required. Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res 1999;17:157-67.) The improved abrasive and adhesive wear resistance that accompanies elevated cross-linking provides other lowered toughness and crack propagation resistance are potential benefits. Backside wear in modular implants, hampered by the lack of good nonlinear material mod- for example, is an important complication that could be els for polyethylene’s stress-strain behavior. Close moni- reduced with the use of elevated cross-linked instead of toring of data from the clinical experience will be conventional polyethylene. Increased wear resistance needed to establish the extent of this problem. also has renewed interest in the use of larger femoral heads for total hip replacement. The risk of dislocation Wear tests of knee-like geometries (with noncon- is markedly reduced with a larger head size. However, forming bearing surfaces) and knee joint simulator stud- because sliding distance between the bearing surfaces ies show that elevated cross-linked materials perform and the amount of wear is higher with a larger head, well in comparison with conventional polyethylene. conventional polyethylene-metal bearings usually have Given their decreased fracture resistance, the hypothesis a small diameter (≤ 32 mm). Larger head sizes are now would be the opposite—wear rates and pitting and available with matching larger diameter, elevated cross- delamination damage should be worse for elevated linked polyethylene acetabular components. The wall cross-linked materials. Lowering of the elastic modulus thicknesses of these components are thin (less than that accompanies the post cross-linking thermal treat- 5 mm in some instances), making the strength and ment used to quench free radicals may explain these toughness of elevated cross-linked polyethylenes an im- positive findings. As in the case of compression-molded portant consideration. polyethylene, the lower modulus creates larger contact areas, lower stresses, and better resistance to wear dam- Changes in mechanical properties that accompany age (Figure 2). increased cross linking may pose the biggest threat to the clinical efficacy of these materials. Reduced tough- ness and resistance to fatigue crack propagation have been shown in several laboratory studies using stan- dardized specimens under controlled loading conditions. These findings suggest greater susceptibility to pitting and delamination wear damage, to gross failure should a crack propagate entirely through a component, and to dissociation that could result from the failure of a lock- ing mechanism that relied on the structural integrity of the polyethylene. The clinical relevance of standardized fracture and fatigue test results is difficult to interpret, because the fracture conditions in implant components depend on geometry and loading conditions that differ from those of standard laboratory specimens. Analytical attempts to interpret performance with consideration of Figure 2 Wear tracks from a conventional polyethylene (1050 resin irradiated at 25 kGy) (A) and an elevated cross-linked polyethylene (1050 resin irradiated at 65 kGy) (B), which were tested on a knee-like wear apparatus for 2 million cycles under identical conditions show more severe damage in the conventional material. The elastic moduli were 1.0 GPa for the conventional and 800 MPa for the elevated cross-linked material. (Reproduced with permission from Furman BD, Maher SA, Morgan T, Wright TM: Elevated crosslinking alone does not explain polyethylene wear resistance, in Kurtz SM, Gsell R, Martell J (eds): Crosslinked and Thermally Treated Ultra-High Molecular Weight Polyethyl- ene for Joint Arthroplasties ASTM STP 1445. West Conshohocken, PA, ASTM International, pp 248-261.) American Academy of Orthopaedic Surgeons 61

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Orthopaedic Knowledge Update 8 Recent findings of increased fatigue crack propaga- mainly for femoral heads in total hip arthroplasties. tion in elevated cross-linked polyethylenes may prove Long-term clinical experience with these materials has problematic for in vivo wear behavior. These results been published. More recently, an oxidized zirconium suggest that if cracks form in vivo after a large number material has been introduced into both hip and knee re- of cycles (larger than the number typically used in labo- placement components for articulation against polyeth- ratory tests), failure might progress rapidly. Thus, even if ylene; however, few results are available and clinical ex- crack initiation is retarded because stresses are lower, perience with this material is short term. the end results could still be less favorable. Long-term experience with alumina-on-polyethylene Ceramics bearings for hip replacement shows reduced wear rates compared with metal-on-polyethylene bearings and an Ceramic materials are solid, inorganic compounds con- associated decrease in the presence of osteolysis, sug- sisting of metallic and nonmetallic elements. Held to- gesting that these types of bearings are beneficial in im- gether by ionic or covalent bonding, ceramics are stiff proving clinical performance. Alumina-on-polyethylene (high elastic modulus), hard, brittle, very strong under bearings for knee replacements have a much more lim- compressive loads, and possess excellent biocompatibil- ited use. Mid-term results are excellent; however, the ity and exceptional wear resistance. Ceramic materials absence of direct comparisons with conventional metal- usually have polygranular microstructures similar to me- on-polyethylene bearing surfaces of the same design tallic alloys. Their properties depend on factors such as and the lack of long-term results make it difficult to as- grain size and porosity (for example, strength is in- sess the clinical benefits. versely proportional to both grain size and porosity). The use of zirconia as a bearing surface against poly- Fully dense ceramics, such as alumina and zirconia, ethylene has not been proven as clinically successful as are used in total joint arthroplasty components specifi- alumina-on-polyethylene bearings. High wear rates, cally because they provide more wear-resistant bearing which resulted in catastrophic failure as early as 5 years surfaces; they have few other mechanical advantages for after implantation, have been reported; however, these joint arthroplasty. Ceramic-on-polyethylene bearings failures may have resulted from the use of an enhanced have been commercially available for some time as al- form of polyethylene (Hylamer, DePuy-Dupont, New- ternatives to conventional metal-on-polyethylene. ark, DE) in the acetabular component, that has been Ceramic-ceramic bearings have only recently received prone to rapid wear. More convincing data exist from a regulatory approval for commercial distribution in the direct comparison between alumina-, zirconia-, and United States. Because of their hardness and strength, metal-on-conventional polyethylene bearings in patients ceramics can be polished to a very smooth finish and re- with total hip arthroplasties. Data show the highest wear sist roughening while in use as a bearing surface. They rate in the zirconia group, which is consistent with an in- possess good wettability, suggesting that lubricating lay- creased monoclinic content on the surface of retrieved ers may form between ceramic couplings, thus reducing zirconia heads from the same series. The propensity for adhesive forms of wear. zirconia to transform from a tetragonal crystalline form, which is stable at elevated temperatures, to the less The most significant disadvantage of ceramics is low tough monoclinic form is a disadvantage of this material toughness, which resulted in a significant number of ce- and has led to FDA warnings against autoclave resteril- ramic head fractures during early clinical use in total hip ization of zirconia heads. The decrease in toughness replacement. More recently, however, improvements in makes the material more susceptible to roughening and ceramic quality, most notably increased chemical purity increased wear. Manufacturing problems led to a high and reduced grain size, has led to a dramatic reduction incidence of fracture and prompted a recent voluntary in head fractures. Toughness is also a concern in the use recall of nine batches of zirconia femoral heads, which of ceramic-on-ceramic acetabular components for hip further undermined confidence in this material. replacements. Retrieval studies of polyethylene acetabu- lar components suggest that impingement between the Ceramic-on-ceramic bearings have been used exten- neck of the femoral component and the rim of the ace- sively in total hip arthroplasty in Europe, though hip im- tabular component is a common occurrence. With the plants using this bearing combination have only recently use of ceramic components, impingement could cause received regulatory approval in the United States. significant damage and eventual fracture. Recent labo- Alumina-on-alumina joints have shown very low clinical ratory tests suggest, however, that with the improved wear rates; however, the results are design dependent quality of ceramic materials the possibility of (these bearings can show excessive wear if incorporated impingement-related failure is quite low. into an inferior design). Recent reports also show excel- lent wear resistance of alumina-on-alumina joints in Three types of ceramic bearing materials are com- young patients with no measurable wear and no evi- mercially available. Bulk implant materials made from dence of osteolysis even at more than 10-year follow-up. alumina and zirconia have been used for decades, Furthermore, head fractures have not been observed in 62 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty this high-demand patient population, which supports the this basic approach include powders that are a blend of supposition that alumina-ceramic materials have im- a copolymer (PMMA and polystyrene or PMMA and proved mechanical properties. Zirconia-on-zirconia methacrylic acid) to increase toughness. bearings have not performed well based on laboratory findings of high wear. The heat given off during polymerization is substan- tial. The rise in temperature that occurs in the adjoining The use of ceramic bearings in other joint arthro- tissues is dictated by the volume and shape of the ce- plasties is more limited in scope. For implant designs ment bulk and the heat transfer properties of the sur- dominated by the potential for abrasive and adhesive rounding structures. Bone cement temperatures during wear (bearing surfaces that remain conforming through- in vivo polymerization are usually below that required out the useful ranges of motion of the implant), ceramic to denature proteins or kill bone cells. The long-term bearing surfaces may have all of the potential benefits successful use of such cement in an array of orthopaedic of ceramic hip replacement components. For joint de- applications shows that the thermal properties do not signs such as conventional knee replacements that re- affect clinical efficacy. quire nonconforming articulations to adequately func- tion, the advantages of ceramic materials are less clear. Antibiotics can be added to bone cement to provide Although knee joint simulator data show substantial de- prophylaxis or treatment of infection. The elution of an- creases in wear rate (for example, an oxidized zirconia tibiotic from the cement is dictated by the preparation surface bearing against polyethylene), the clinical bene- technique, chemistry, and surface area of the cement. fit of this decreased wear rate remains uncertain. Os- The elution of gentamycin from commercially available teolysis has not been as significant a complication in cements varies significantly depending on the brand knee replacement as in hip replacement surgery; long- used. Bone cement properties can be detrimentally al- term clinical data will be required to substantiate a clear tered by the addition of antibiotics during the mixing advantage. Fatigue-related pitting and delamination is process, an important clinical concern when antibiotics the main wear problem after knee replacement. This are added to the cement at surgery. The FDA has re- type of wear is caused by the large cyclic stresses gener- cently approved commercially prepared antibiotic ce- ated at and near the polyethylene surface of the tibial ments for use in the United States; this will markedly component as the more rigid femoral component moves reduce potential complications. across it. Ceramic materials provide no advantage in such a situation. The performance of cement as a grout for fixation of joint arthroplasty components has been enhanced by Bulk ceramic components for knee replacement have improved protocols in cement handling, bone prepara- been used primarily in Japan, where both ceramic-on- tion, and cement delivery. For example, vacuum-mixed polyethylene and ceramic-on-ceramic bearings have been or low viscosity formulations of cement that are deliv- used clinically. Evidence of the efficacy of either bearing ered into the bone under pressure using a cement gun combination is limited by the length of follow-up (< 9 have a significantly reduced porosity compared with years), the lack of comparative data for other bearing sur- hand-mixed cement. Reducing the porosity results in faces with the same design, and patient selection (for ex- more cement in the mantle and increased structural ample, one study was limited to patients with rheumatoid strength, even though the material properties of the ce- arthritis). Whereas satisfactory outcomes have resulted ment itself remain unaltered. Although difficult to prove with the use of ceramic bearings in total knee arthroplasty, clinically, a reduction in porosity should reduce the no data yet exist to prove them superior to the more com- chances of cement mantle fracture and subsequent im- mon metal-on-polyethylene articulations. plant loosening. Cements PMMA cement has often been used solely as a structural component to fill bone defects and as a load- Polymethylmethacrylate (PMMA) bone cement has carrying element of the resulting composite. Vertebro- been the polymer of choice as a grouting agent to se- plasty and kyphoplasty are gaining acceptance as treat- cure implant components to bone since its introduction ments for painful osteoporotic compression fractures. by Charnley in the 1970s. The basic principles of the in Vertebroplasty was first used for osteolytic spinal condi- situ polymerization remain the same. Liquid methyl- tions secondary to cancer; injection of cement into the methacrylate monomer with the addition of hydro- vertebral body can prevent further collapse and relieve quinone (to inhibit premature polymerization) and pain. Kyphoplasty, an extension of vertebroplasty, uses N,N,-dimethyl-P-toluidine (to accelerate polymerization an inflatable bone tamp to restore the vertebral height once mixing commences) is mixed with prepolymerized while creating a cavity to be filled with bone cement. PMMA, which also contains dibenzoyl peroxide (to ini- Short-term clinical results with both procedures are tiate the polymerization process) and a radiopaque ma- promising; however, several biomechanical questions re- terial (usually barium sulfate or zirconia). Variations on main, including the appropriate properties for the ce- ment, the amount of fill, and the effects of vertebra rein- forcement on the adjacent spinal motion segments. American Academy of Orthopaedic Surgeons 63

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Orthopaedic Knowledge Update 8 Serious complications related to cement leakage, includ- ical properties that can be varied through altering the ing soft-tissue damage and nerve root pain and com- material’s composition and structure. Hydrogels have pression, have prompted the FDA to issue a public been considered for use in a wide range of biomedical health notification concerning this off-label use of ce- and pharmaceutical applications; orthopaedic applica- ment. tions include tissue engineering of cartilage and bone and for drug delivery. Interest in using bioactive cements in these types of applications is growing. Cements based on hydroxyapa- Interfaces tite and calcium phosphate have been studied to deter- mine their ability to provide strength and stiffness to os- In orthopaedic applications, biomaterials are generally teoporotic vertebral bodies compared to that achieved used to fabricate entire structures for specific purposes with PMMA. Handling and radiopacity difficulties have (for example, a fracture plate or a tibial knee replace- hampered the clinical use of these cements in spinal ap- ment insert). Difficulties in achieving long-term perma- plications. These cements have been found to be effec- nent fixation of orthopaedic devices to the skeleton has tive as augments to standard fixation in several frac- led to the development of specific materials and tech- tures, including those of the distal radius and femoral nologies intended to enhance biologic fixation. For ex- neck. ample, tantalum is a highly biocompatible, corrosion- resistant, osteoconductive material. Recently, porous Biodegradable Polymers forms of tantalum deposited on pyrolytic carbon back- bones have been suggested as superior structures for Biodegradable polymers degrade chemically in a con- bone ingrowth. Orthopaedic applications include coat- trolled manner over time. Orthopaedic applications for ings for joint arthroplasty components (acetabular cups these materials include sutures, screws, anchors, and pins and tibial trays) and as spinal cages. Experimental work designed to slowly lose their mechanical function as in animal models and randomized trials in humans sug- they resorb and the surrounding tissue heals. The tissue gest that this may be a useful material for achieving fix- assumes its normal mechanical role, while the resorption ation to bone. of the polymer eliminates the need for a second surgical procedure to remove the device. Resorbable polymers Hydroxyapatite, a ceramic, has been available as a also are used for drug delivery, and considerable re- coating on joint arthroplasty implants for some time. search is underway to develop biodegradable scaffolds Animal studies showed that such coatings increase fixa- for tissue engineering. This last application is especially tion strength by preferential deposition of new bone at challenging because the scaffold must provide a suitable the interface. Despite this in vivo evidence, no clear ad- biologic environment for the cells to be delivered within vantage has been shown in the clinical literature; the material and a suitable mechanical environment so whereas some reports suggest enhanced fixation, others that they manufacture extracellular matrix with the ap- show no benefit. This disagreement may be the result of propriate biomechanical properties. Biocompatibility is several factors, including patient selection, implant de- a very important consideration; the polymer must de- sign, and the quality and type of hydroxyapatite used. grade without adversely affecting the cells or the tissues that replace it. Summary Common bioresorbable polymers include polylactic A clinical improvement in wear rate is possible with the acid, polyglycolic acid, polydioxanone, and polycaprolac- introduction of alternative bearing materials. In all tone. Their mechanical properties span large ranges, de- cases, however, compromises exist; although wear be- pending on polymer type, the addition of copolymers, havior can be improved, other properties are altered in molecular weight, fabrication technique, and the addi- potentially detrimental ways (Table 1). The use of pre- tion of reinforcing materials such as fibers. Orthopaedic clinical assessment tools such as joint simulators are vi- applications are limited by the strengths of these materi- tal to establishing efficacy, but the shortcomings of these als, which are insufficient for many load-bearing situa- tools must be considered when interpreting the clinical tions common in the musculoskeletal system. Bioresorb- relevance of test results. Long-term follow-up data from able polymers are most often used in applications well-controlled studies remain the only real test of effi- involving trauma; such as the fixation of small cancel- cacy. lous bone fractures. The most common use reported in the literature is in association with malleolar fractures. Annotated Bibliography Hydrogels are a soft, porous-permeable group of General Reference polymers that are nontoxic, nonirritating, nonmutagenic, nonallergenic, and biocompatible. They readily absorb McKellop HA: Bearing surfaces in total hip replace- water (and thus have high water contents), and have ments: State of the art and future developments. Instr low coefficients of friction and time-dependent mechan- Course Lect 2001;50:165-179. 64 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty This review article discusses the advantages and disadvan- smaller heads and significantly higher wear rates compared tages of bearing surfaces including the new polyethylenes, with those found with larger diameter heads. In the larger modern metal-metal, and ceramic-ceramic bearings. The goal heads, surface separation occurred, suggesting the formation of this review is to provide surgeons with the information of a protective lubricating film. needed to assess the risk-benefit ratios of each of the new bearing combinations. Ultra-High Molecular Weight Polyethylene Santavirta S, Bohler M, Harris WH, et al: Alternative Burroughs BR, Rubash HE, Harris WH: Femoral head materials to improve total hip replacement tribology. sizes larger than 32 mm against highly cross-linked poly- Acta Orthop Scand 2003;74:380-388. ethylene. Clin Orthop 2002;405:150-157. This review article examines proposed improvements in In vitro wear testing and anatomic studies, together with tribology of bearing surfaces for total hip replacement. The prior clinical studies in which large femoral heads were used, three approaches examined were highly cross-linked support the hypothesis that highly cross-linked UHMWPE al- UHMWPE, aluminum oxide, and metal-on-metal. The findings lows the use of femoral heads larger than 32 mm. in support of their efficacy were emphasized. Collier JP, Currier BH, Kennedy FE, et al: Comparison Wright TM, Goodman SB (eds): Implant Wear in Total of cross-linked polyethylene materials for orthopaedic Joint Replacement: Clinical and Biologic Issues, Materi- applications. Clin Orthop 2003;414:289-304. als and Design Considerations. Rosemont, IL, American Academy of Orthopaedic Surgeons, 2001. Physical and mechanical properties of six commercially available cross-linked polyethylene materials were obtained This monograph provides answers to important clinical, bi- from tests in a single laboratory and reported along with wear ologic, and bioengineering questions concerning the complica- data and summary descriptions of the materials by the manu- tions caused by wear in total joint arthroplasties. The answers facturers. were drafted by invited attendees at a workshop sponsored by the National Institutes of Health and the American Academy Digas G, Karrholm J, Thanner J, Malchau H, Herberts P: of Orthopaedic Surgeons and are supported by a comprehen- Highly cross-linked polyethylene in cemented THA: sive bibliography. Randomized study of 61 hips. Clin Orthop 2003;417:126- 138. Metal-on-Metal Bearings Highly cross-linked polyethylene components (irradiated Amstutz HC, Beaule PE, Dorey FJ, Le Duff MJ, Camp- to 95 kGy) were compared with conventional polyethylene bell PA, Gruen TA: Metal-on-metal hybrid surface ar- components in a randomized study of cemented hip replace- throplasty: Two to six-year follow-up study. J Bone Joint ments. At 2-year follow-up, the highly cross-linked cups Surg Am 2004;86:28-39. showed 50% reduction of proximal wear based on radiostere- ometric examinations; however, the follow-up period was Encouraging results were found in this follow-up study short and the results preliminary. (average follow-up 3.5 years) of 400 hips treated with metal- on-metal surface arthroplasties. Survival rates of the arthro- D’Lima DD, Hermida JC, Chen PC, Colwell CW Jr: plasties at 4 years were 94.4%; however, a 3% revision rate for Polyethylene cross-linking by two different methods re- loosening of the femoral component and femoral neck frac- duces acetabular liner wear in a hip joint wear simula- ture were also found. tor. J Orthop Res 2003;21:761-766. Goldberg JR, Gilbert JL, Jacobs JJ, Bauer TW, Paprosky A comparison of wear reduction by two different cross- W, Leurgans S: Links: A multicenter retrieval study of linking methods, 9.5 Mrad electron beam and 10 Mrad gamma the taper interfaces of modular hip prostheses. Clin irradiation, showed no clinically significant differences. Both Orthop 2002;401:149-161. methods were superior to conventional nominally cross-linked (gamma sterilized) polyethylene in wear resistance. This study found that corrosion and fretting of modular taper surfaces was significantly worse for mixed alloy couples Endo MM, Barbour PS, Barton DC, et al: Comparative compared with similar alloy couples and were also dependant wear and wear debris under three different counterface on implantation time and flexural rigidity of the neck. The re- conditions of crosslinked and non-crosslinked ultra high sults suggest that this damage was attributable to a mechani- molecular weight polyethylene. Biomed Mater Eng cally assisted, crevice corrosion process. 2001;11:23-35. Smith SL, Dowson D, Goldsmith AA: The effect of fem- The wear of grade GUR 1020 cross-linked, GUR 1120 oral head diameter upon lubrication and wear of metal- cross-linked, and non–cross-linked GUR 1020 UHMWPE on-metal total hip replacements. Proc Inst Mech Eng (Ticona, Summit, NJ) was compared in multidirectional wear [H] 2001;215:161-170. tests using three different counterface conditions. Better wear resistance was found with cross-linking for smooth counter- A hip joint simulator study of metal-on-metal joints with faces, but with rough and scratched counterfaces, the GUR different head diameters showed no surface separation with American Academy of Orthopaedic Surgeons 65

Biomaterials and Bearing Surfaces in Total Joint Arthroplasty Orthopaedic Knowledge Update 8 1020 cross-linked material produced significantly higher wear more severe wear. No change in wear rate occurred in the alu- rates than the non–cross-linked material. mina group, which showed considerably less wear. Hopper RH Jr, Young AM, Orishimo KF, Engh CA Jr: Urban JA, Garvin KL, Boese CK, et al: Ceramic-on- Effect of terminal sterilization with gas plasma or polyethylene bearing surfaces in total hip arthroplasty: gamma radiation on wear of polyethylene liners. J Bone Seventeen to twenty-one-year results. J Bone Joint Surg Joint Surg Am 2003;85:464-468. Am 2001;83:1688-1694. Patients who received hip replacements with polyethylene Findings showed that long-term use of ceramic-on- acetabular liners sterilized with gamma radiation (61 hips at polyethylene bearings implanted with cement by one surgeon mean follow-up of 5.2 years) experienced an average of 50% resulted in outstanding long-term clinical and radiographic re- less wear than patients with non–cross-linked liners sterilized sults and wear rates lower than those previously reported for with gas plasma (63 hips at mean follow-up of 3.9 years). metal-on-polyethylene bearings. These findings support the use of such bearings in total hip arthroplasty. Muratoglu OK, Bragdon CR, O’Connor DO, Perinchief RS, Jasty M, Harris WH: Aggressive wear testing of a Cements cross-linked polyethylene in total knee arthroplasty. Clin Orthop 2002;404:89-95. Cassidy C, Jupiter JB, Cohen M, et al: Norian SRS ce- ment compared with conventional fixation in distal ra- A knee simulator study comparing elevated cross-linked dial fractures: A randomized study. J Bone Joint Surg and conventional tibial inserts showed extensive delamination Am 2003;85:2127-2137. and cracking as early as 50,000 cycles for conventional poly- ethylene inserts, whereas the cross-linked polyethylene inserts A prospective, randomized multicenter study of closed re- did not show any subsurface cracking or delamination at duction and immobilization with and without the use of cal- 0.5 million cycles. cium phosphate bone cement for the treatment of distal radial fractures, found significant clinical improvement during the Ceramics first 3 months of treatment with cement augmentation. Allain J, Roudot-Thoraval F, Delecrin J, Anract P, Mi- Garfin SR, Reilley MA: Minimally invasive treatment of gaud H, Goutallier D: Revision total hip arthroplasty osteoporotic vertebral body compression fractures. performed after fracture of a ceramic femoral head: A Spine J 2002;2:76-80. multicenter survivorship study. J Bone Joint Surg Am 2003;85:825-830. A literature review with presentation of early results of a national, clinical study showed 95% of patients treated for One hundred five surgical revisions to treat a fracture of a painful osteoporotic compression fractures had significant im- ceramic femoral head, performed at 35 institutions, were stud- provement in symptoms and function after kyphoplasty or ver- ied. Although the fractures were potentially serious events, tebroplasty. Only kyphoplasty improved vertebral height and treatment with total synovectomy, cup exchange, and insertion kyphosis. of a cobalt-chromium or new ceramic femoral ball minimized the risk of early loosening and the need for one or more re- Lewis G: Fatigue testing and performance of acrylic peat revisions. bone-cement materials: State-of-the-art review. J Biomed Mater Res 2003;66B:457-486. Hamadouche M, Boutin P, Daussange J, Bolander ME, Sedel L: Alumina-on-alumina total hip arthroplasty: A A literature review of the fatigue behavior of acrylic bone- minimum 18.5-year follow-up study. J Bone Joint Surg cement formulations showed that an increase in molecular Am 2002;84:69-77. weight leads to an increase in fatigue life; the effect of vacuum mixing on fatigue life remains controversial. A retrospective long-term study showed a 20-year survival rate of 85% for alumina-on-alumina hip replacements with Biodegradable Polymers minimal wear rates and limited osteolysis, provided that the acetabular component remained well-fixed. No alumina head Grande DA, Mason J, Light E, Dines D: Stem cells as or socket fractures occurred and osteolysis was moderate in platforms for delivery of genes to enhance cartilage re- about 10% of the patients. pair. J Bone Joint Surg Am 2003;85(suppl 2):111-116. Hernigou P, Bahrami T: Zirconia and alumina ceramics Stem cells transduced with either bone morphogenetic in comparison with stainless-steel heads: Polyethylene protein-7 or sonic hedgehog gene were delivered to osteo- wear after a minimum ten-year follow-up. J Bone Joint chondral defects in rabbits using bioresorbable scaffolds. The Surg Br 2003;85:504-509. addition of either factor enhanced the quality of the repaired tissue, showing the utility of tissue-engineering gene therapy Comparison of ceramic and metal bearing surfaces showed strategies. increased polyethylene wear at 5- to 12-year follow-up for zir- conia and metal heads. The group with the zirconia heads had Hovis WD, Kaiser BW, Watson JT, Bucholz RW: Treat- ment of syndesmotic disruptions of the ankle with bio- 66 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 5 Biomaterials and Bearing Surfaces in Total Joint Arthroplasty absorbable screw fixation. J Bone Joint Surg Am 2002; Boehler M, Plenk H Jr, Salzer M: Alumina ceramic 84:26-31. bearings for hip endoprostheses: The Austrian experi- ences. Clin Orthop 2000;379:85-93. Syndesmotic disruptions associated with malleolar frac- tures in which the syndesmosis was fixed with poly-L-lactic Chan FW, Bobyn JD, Medley JB, Krygier JJ, Tanzer M: acid screws healed without displacement of the syndesmosis or The Otto Aufranc Award. Wear and lubrication of widening of the medial clear space. No episodes of osteolysis metal-on-metal hip implants. Clin Orthop 1999;369: or late inflammation secondary to the hydrolyzed polylactide 10-24. occurred. Dorr LD, Wan Z, Longjohn DB, Dubois B, Murken R: Interfaces Total hip arthroplasty with use of the Metasul metal-on- metal articulation: Four to seven-year results. J Bone Kim YH, Kim JS, Oh SH, Kim JM: Comparison of Joint Surg Am 2000;82:789-798. porous-coated titanium femoral stems with and without hydroxyapatite coating. J Bone Joint Surg Am 2003;85: Dorr LD, Wan Z, Song M, Ranawat A: Bilateral total 1682-1688. hip arthroplasty comparing hydroxyapatite coating to porous-coated fixation. J Arthroplasty 1998;13:729-736. At a mean follow-up of 6.6 years postoperatively, the clini- cal and radiographic results associated with proximally Fenollosa J, Seminario P, Montijano C: Ceramic hip porous-coated femoral prostheses with identical geometries prostheses in young patients: A retrospective study of 74 that differed only with regard to the presence or absence of patients. Clin Orthop 2000;379:55-67. hydroxyapatite coating were found to be similar. Wigfield C, Robertson J, Gill S, Nelson R: Clinical expe- Jacobs JJ, Skipor AK, Patterson LM, et al: Metal release rience with porous tantalum cervical interbody implants in patients who have had a primary total hip arthroplas- in a prospective randomized controlled trial. Br J ty: A prospective, controlled, longitudinal study. J Bone Neurosurg 2003;17:418-425. Joint Surg Am 1998;80:1447-1458. A prospective randomized study of the ability of porous McKellop H, Shen FW, Lu B, Campbell P, Salovey R: tantalum implants to achieve cervical interbody fusion showed Development of an extremely wear-resistant ultra high inferior end-plate lucency in four patients raising concerns molecular weight polyethylene for total hip replace- about delayed or nonfusion. Fusion subsequently occurred in ments. J Orthop Res 1999;17:157-167. all patients at 12 months after surgery. Oonishi H, Kadoya Y: Wear of high-dose gamma- Classic Bibliography irradiated polyethylene in total hip replacements. J Orthop Sci 2000;5:223-228. Bartel DL, Rawlinson JJ, Burstein AH, Ranawat CS, Flynn WF Jr: Stresses in polyethylene components of Ritter MA: Direct compression molded polyethylene contemporary total knee replacements. Clin Orthop for total hip and knee replacements. Clin Orthop 2001; 1995;317:76-82. 393:94-100. Bobyn JD, Toh KK, Hacking SA, Tanzer M, Krygier JJ: Rokkanen PU, Bostman O, Hirvensalo E, et al: Bioab- Tissue response to porous tantalum acetabular cups: A sorbable fixation in orthopaedic surgery and traumatol- canine model. J Arthroplasty 1999;14:347-354. ogy. Biomaterials 2000;21:2607-2613. American Academy of Orthopaedic Surgeons 67

6Chapter Physiology of Aging Susan V. Bukata, MD Mathias Bostrom, MD Joseph A. Buckwalter, MD Joseph M. Lane, MD Introduction glia occurs with a decrease in the total number of neu- rons. A decrease in brain tissue metabolism accompa- Aging is a complex process that involves changes in nies a decrease in cerebral blood flow. Processing in many of the physiologic functions of the body; aging both nervous systems slows with increasing age, with also may create changes in social and economic condi- nerve conduction velocity decreased 10% to 15%. Auto- tions that influence lifestyles and daily routines. The ac- nomic and muscle stretch reflexes become less sensitive cumulated effects of past diseases, side effects of medi- and righting reflexes decrease in acuity, resulting in in- cations, and environmental factors can change an creased body sway that makes it more difficult for eld- individual’s ability to perform daily activities. Older erly people to respond to sudden positional changes. people often have difficulty sleeping and many experi- Reaction times in older adults are 20% longer than ence at least some recent memory loss. The heterogene- those in young adults, which partially accounts for the ity of the aging body also makes patient response to a 35% to 40% increase in falls seen in adults older than particular medication or treatment much more variable 60 years. than in a younger population. To care for the elderly pa- tient and improve quality of life, it is important to un- An age-related decrease in the number of spinal mo- derstand the changes that occur with aging and to ad- tor neurons probably contributes to these changes. Sin- dress the special needs that these changes produce. The gle peripheral motor neurons innervate groups of skele- United States Census Bureau projects that by the year tal muscle fibers forming a motor unit. Accompanying 2010, one quarter of the US population will be age 55 the loss of spinal motor neurons is a concomitant loss in years or older. This projection is based in part on in- the total number of motor units in old muscles and re- creasing life expectancies for men and women. Health modeling of other motor units. This decrease in total care providers must understand the special needs associ- motor units is actually a specific loss of fast motor units ated with aging to provide appropriate treatment for and an increase in slow motor units. During normal ag- this expanding population of patients. ing, a reorganization of the motor unit pool for skeletal muscle occurs, which supports the belief that some fast Changes in Neuromuscular Control fibers may undergo denervation and others may be rein- nervated by sprouting nerves from the slow motor units. The maintenance of a moderate level of neuromuscular Changes also occur at the neuromuscular junction, control is essential for the continuation of normal daily which affect the recruitment of muscle fibers for coordi- routines. Many changes that occur with natural aging af- nated activity. With aging, there is a degeneration of the fect neuromuscular control. These changes affect an in- neuromuscular junction that prevents or slows the trans- dividual’s ability to perform certain tasks and result in mission of neural stimuli to muscle fibers. All of these changes in proprioception that affect balance and in- changes contribute to slower reaction times. crease the risk for falling. Changes in proprioception stem from a variety of sources, such as sensory changes Many sensory changes also occur with aging that can that affect vision, hearing, and vestibular function, as make it difficult for eldery people to respond to changes well as age-related changes in the central and peripheral in their environment. Increasingly poor vision, macular nervous systems. All of these changes affect coordina- degeneration, stereopsis, cataracts, and poor night vision tion and musculoskeletal function by altering the neural commonly occur with aging. Hearing loss with a de- control and stimuli of muscles, which affect mobility. crease in high-frequency acuity (affecting the ability to distinguish words from background noise) affects almost Many other changes occur in the central and periph- 30% of patients older than 70 years and almost 90% of eral nervous systems with aging. Brain atrophy of 30% nursing home residents. Impaired vestibular function to 40% in the frontal lobe, temporal lobe, and basal gan- also affects many older patients and may play a signifi- American Academy of Orthopaedic Surgeons 69

Physiology of Aging Orthopaedic Knowledge Update 8 Figure 1 A, Osteoporotic trabecular bone. B, Normal trabecular bone. (Adapted with creases 40% and the total number of fibers declines permission from Dempster DW, Shane E, Horbert W, Lindsay R: A simple method for 39% between the ages of 20 and 80 years. In addition, correlative light and scanning electron microscopy of human iliac crest bone biopsies: greater muscle mass is lost in weight-bearing muscles Qualitative observations in normal and osteoporotic subjects. J Bone Miner Res 1986; than in non–weight-bearing muscles. These combined 1:15-21.) changes in muscle mass lead to a prolongation of twitch contraction and a reduction in voluntary strength. Poor cant role in balance impairment and falls. Neuropathy quadriceps function has been identified as a key risk secondary to diabetes and nerve injuries can limit sensa- factor for falls and hip fracture. tion in the feet and impair the ability to respond to un- even terrain. Impaired cognition secondary to medica- Aging also causes many significant changes in the tions, dementia, depression, or stroke also can impair an cardiovascular system, such as changes in functional re- older person’s ability to respond appropriately to their serve capacity that affect an individual’s ability to re- environment and is a risk factor for appendicular frac- spond to physiologic stresses. Increased vascular stiff- ture. ness, hypertension, and other factors result in a decreased ability to increase cardiac output in response Changes in Body Composition and to physiologic stress. Maximal oxygen consumption, a Functional Reserve measurement that serves as a good measure of athletic fitness, declines 5% to 15% each decade after age 25 Age-related changes occur in body composition that re- years. Endurance training is able to slow this decline up sult in decreased bone mass, decreased muscle mass, and to 50% until approximately age 70 years. Older patients an increase in body fat. Sarcopenia, or muscle mass loss, are able to respond to exercise and to improve muscle reduces basal metabolism and is the predominant cause strength and tone and cardiovascular response; how- of decreased muscle strength seen with aging. This loss ever, even very fit older adults may have a maximal of strength can have a profound effect on an elderly functional reserve that only is equivalent to that of person’s ability to perform simple tasks such as walking younger sedentary individuals. and rising from a chair. Peak muscle mass is attained at approximately 30 years of age. After age 50 years, ap- Bone mass also changes as a part of normal aging. proximately 15% of muscle mass is lost per decade, in- Peak bone mineral density is reached between age 25 creasing up to 30% per decade after age 70 years. Mus- and 30 years and begins to decline in the fourth and cle atrophy occurs from both a reduction in fiber size fifth decades of life. Bone mineral density declines 0.3% and quantity. The equal loss of type I (fast twitch) and to 0.5% per year and the rate of loss may increase for type II (slow twitch) fibers accounts for most of the at- women to 2% to 3% per year in the 7- to 10-year period rophy associated with aging; however, the type II fibers of perimenopause. The rate of bone loss and the associ- also decrease in size more than the type I fibers and ated increase in fracture risk is related to a complex subsequently compose a smaller percentage of the mus- combination of genetics, body mass characteristics, nutri- cle cross-sectional area. A 26% reduction in the size of tional factors, and diseases. A complex relationship, type II fibers occurs from age 20 to 80 years and ac- commonly referred to as coupling, exists between osteo- counts for a large proportion of age-related muscle mass blasts and osteoclasts, which balances bone formation loss. This decrease in fiber size is not seen with normal with bone resorption. It is ultimately this relationship disuse atrophy. Overall muscle cross-sectional area de- between the osteoblasts and osteoclasts that determines the rate of bone mineral density loss. After peak bone mass is attained, the full bone remodeling unit is not completely replaced during the process of normal bone turnover, which results, over time, in decreased bone mass. During the perimenopausal period, the activity of osteoclasts is enhanced and osteoblasts are unable to re- place bone at the same pace. Architectural changes eventually occur as bone trabeculae decrease in number and lose continuity. Whereas cortical bone thins, the losses are greater in the trabecular bone mass (Figure 1). More information about bone mineral density and bone remodeling can be found in chapter 18. All of these processes combine to increase the fragility of ag- ing bone. Fracture repair is not delayed or compro- mised, but fixation failure is more common because of inadequate bone stock. Almost 98% of elderly patients who sustain a hip fracture have some level of bone defi- ciency. Low-energy hip fractures are associated with a 70 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 6 Physiology of Aging 20% increase in mortality within the first 6 months after trix composition, and molecular organization. Localized, injury. Although patients with two or more vertebral superficial fibrillation of articular surfaces can be first fractures have a 25% to 30% increase in mortality com- seen in adolescents as they approach skeletal maturity. pared with peers who do not sustain such fractures, this During skeletal growth, there is a noted decrease in chon- difference is reflected over the much longer period of drocyte density, an associated decrease in synthetic activ- 5 years. Because of age-associated bone loss, elderly ity, and a change in the proteoglycan component ratios. people are at greater risk for low-energy fractures and With increasing age, there is a decrease in the ratio of related mortality. chondroitin sulfate to total glycosaminoglycans, an in- crease in the production of chondroitin 6 sulfate, and a de- Normal aging results in many changes in the endo- crease in the production of chondroitin 4 sulfate. There crine system that affect a variety of body tissues and also is a decreased response to growth factors, especially functions. Changes in glucose metabolism include an in- transforming growth factor-β, which exerts crease in peripheral glucose resistance and an increase almost no influence on chondrocytes once skeletal matu- in postprandial blood glucose and insulin levels. These rity is reached, but is a strong stimulant to glycosaminogly- changes are influenced by diet, exercise, and body fat can production (particularly chondroitin 4 sulfate). There content, all of which often change with aging. This al- is an age-associated decrease in the water content of car- tered glucose metabolism results in a higher rate of tilage, in direct contrast to the changes that occur with de- insulin-resistant diabetes (type II diabetes) in the older generative arthritis. In osteoarthritis, water content ini- population. Decreased levels of growth hormone, insulin tially increases as a result of a disruption in the growth factor, estrogen, and testosterone also occur with architecture of the matrix molecules. Other matrix aging. Patients have been treated with growth hormone changes associated with aging include an increase in deco- supplementation, but the benefits are not clear. Estro- rin concentration, collagen cross-linking, collagen fibril di- gen supplementation was used for many years as the ameter, and variability in collagen fibril diameter. mainstay of treatment for patients with perimenopausal bone loss; however, global risk including increases in the Chondrocyte proliferation and cell density begin to prevalence of breast cancer, strokes, pulmonary emboli, decline from birth, and reach a steady plateau by skele- and blood clots are believed to outweigh the benefits tal maturity. Although cell number may not vary signifi- obtained in fracture reduction and decreased bone loss. cantly with increasing age, the synthetic capabilities and Estrogen supplementation is no longer recommended as cell morphology continue to change. Cells accumulate a treatment if its sole purpose is to prevent bone loss. intracytoplasmic filaments and may lose some of their endoplasmic reticulum. Cell culture studies using chon- Age-Related Changes in Articular Cartilage drocyte samples extracted from human articular carti- lage show a decrease in absolute chondrocyte yield and Most of the cells and extracellular matrix proteins of ar- in cell proliferation after the age of 30 years. These cell ticular cartilage undergo very little turnover, which re- cultures are responsive to platelet-derived growth fac- sults in an accumulation of age-related changes and tor–BB, with a threefold increase in glycosaminoglycan damage to the tissue over many years. In the world pop- production up to age 40 years; there is minimal response ulation, almost 25% of people older than 60 years have after that age. A recent study showed that human articu- osteoarthritis, which makes age the greatest risk factor lar cartilage chondrocytes become senescent with in- for developing this disease. However, the relationship creasing patient age, which may account for the de- between age and osteoarthritis is not well understood. creased capacity of chondrocytes to respond to injury. In The age-related changes in articular cartilage are some- addition, oxidative stress to the cartilage tissue may what different from those seen in osteoarthritis; how- cause premature chondrocyte senescence, making the ever, these changes affect the biomechanical, biochemi- tissue susceptible to injury for a longer period of time. cal, and cellular characteristics of the tissue and possibly Chondrocyte cultures maintained in 5% oxygen dou- increase the risk of damage and progression to osteoar- bled 60 times before becoming senescent, whereas in- thritis. The extent of degenerative changes varies by creasing the oxygen concentration to 21% cut the num- joint; however, not all degenerative changes are associ- ber of population doublings to 40. Increased oxidative ated with joint pain and loss of motion. All degenerative stress occurs after trauma and with inflammation and changes do not progress to osteoarthritis, which further may explain why injury increases a patient’s risk for os- indicates the complex relationship between the two pro- teoarthritis. cesses. Age-related changes to the synovium, subchon- dral bone, menisci, ligaments, and the joint capsule may Mechanical studies have shown significant age- all contribute to the progression of joint symptoms and related changes in the tensile properties of articular car- the establishment of osteoarthritis. tilage including decreases in stiffness, fatigue resistance, and strength. The exact cause of these changes is not Articular cartilage undergoes age-related changes in known but it is likely that changes in the collagen ma- thickness, cellular function, matrix tensile properties, ma- trix, the proteoglycans, and the decrease in water con- American Academy of Orthopaedic Surgeons 71

Physiology of Aging Orthopaedic Knowledge Update 8 tent all play a role. The increase in advanced glycosyla- clines and the remaining collagen is composed of fibrils tion end products (AGEs) and their effect on cartilage with increased diameter and diameter variability, and mechanical properties are of significant interest. AGEs decreased pyridinoline crosslinks. In later adulthood, adversely influence cartilage turnover, causing decreases there is increased scar formation within the disk (includ- in matrix synthesis and degradation. Their presence is ing the anulus fibrosus), a disruption in the normal accompanied by increased cartilage stiffness and brittle- lamellar pattern of collagen fibrillar layers within the ness, possibly because of increased cross-linking of the anulus fibrosus, and a decrease in the functional stretch collagen molecules by the AGEs. A canine model of an- of these collagen molecules. In elderly people, it be- terior cruciate ligament (ACL) transaction showed in- comes difficult to distinguish the anatomic regions of creased collagen damage, impaired matrix repair, de- the disk. creased proteoglycan synthesis, and more severe osteoarthritis in the animals treated to have a higher Significant alterations in the blood supply to the disk concentration of AGEs in their tissues, similar to that appear to parallel the structural changes. With increasing found in older dogs. If there is confirmation that these age, physiologic vessels within the disk disappear. The os- findings are correlated with the onset of osteoarthritis, sification of the vertebral end plates that occurs with skel- reversing or preventing the changes modulated by etal maturity may significantly affect the nutritional sup- AGEs may represent a novel approach to preventing ply and hydration of the adjacent disk.The cells within the the onset of osteoarthritis. Collagen fibrils also become disk begin to depend on diffusion of nutrients from the larger in diameter and more variable in size with aging; periphery, a process that appears to be hindered as the this change is attributed to a decrease in the type XI number of vessels supplying the disk declines and in- collagen component. The larger, more cross-linked creased intradiskal matrix slows diffusion through the tis- fibrils are more rigid and may limit the ability of the ar- sues. Decreased hydration of the central nucleus pulposus ticular cartilage surface to deform without cracking. leads to stiffening of the surrounding anulus fibrosus and changes in spine biomechanics that can affect load shar- Age-Related Changes in Intervertebral Disk ing between the adjacent facet joints, ligaments, paraspi- nal muscles, and vertebral bodies. Other diseases and lif- Low back pain, which affects a substantial portion of estyle habits that affect small vessel proliferation and the adult population, is one of the most common rea- function, including diabetes and smoking, may accelerate sons for an adult to consult a primary care physician. Al- the rate of change in the disk tissues. though it is difficult to separate age-related changes from degenerative changes to the intervertebral disk, it Age-Related Changes in Tendon, Ligament, is clear that substantial changes occur (possibly more and Joint Capsule than occur to any other musculoskeletal tissue) as a re- sult of aging. There are also substantial individual differ- Injuries to ligaments, tendons, and joint capsules are ences in the rate that these changes occur, which may commonly seen in middle-aged and elderly individuals. explain the difficulty in separating the effects of aging These injuries can result in significant impairment in from those of degeneration. Several general trends link strength, range of motion, and stability in the regions the blood supply, cellular and matrix proliferation, and where they occur. Significant pain and disability can re- tissue hydration to the structural changes observed in sult from low-energy ruptures of the rotator cuff, quad- the intervertebral disk. riceps, patellar, bicep, or posterior tibial tendons, and form ruptures to spinal and wrist ligaments following Beginning at birth, the abundant notochordal cells physical activity or even as a result of the normal activi- found in the nucleus pulposus begin to disappear, ties of daily living. whereas chondrocyte density increases. The cellular den- sity at the disk end plates begins to increase and disor- Although the cellular and mechanical properties that ganization of the cartilaginous regions is noted. By the change with aging have not been fully defined in all of second decade of life, blood vessels within the disk be- these tissues, similarities in structure and composition gin to disappear and cleft formation in the nucleus pul- suggest that the known observations are probably appli- posus begins. During adulthood, the number of clefts cable to all of these tissues. Fibroblast cells form a major continues to increase and radial tears develop that ex- proportion of ligaments, tendons, and joint capsules. As tend into the disk periphery and the anulus fibrosus. fibroblasts age, the cells flatten, elongate, and lose most There is a marked decline in the number of cells within of their protein-producing cell organelles, including the the disk, and the rate of cellular apoptosis increases. rough endoplasmic reticulum and Golgi apparatus. This Both the volume and shape of the disk change. Mucoid finding suggests that the tissues are less biosynthetically degeneration with an increase in noncollagenous pro- active, and may be less able to efficiently respond to in- tein content occurs and there is a concomitant decrease jury. Several factors may contribute to the age-related in proteoglycan and water content. Collagen content de- increase in injuries to these tissues. A decreasing blood supply may cause changes in cell activity and matrix 72 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 6 Physiology of Aging composition and organization that lead to tissue degen- meals, supplemental nutrition (such as shakes and pud- eration. Although biochemical analysis of these tissues dings) should be added to their daily diet. has not identified dramatic changes in the matrix com- position, a decrease in collagen cross-links (common in Older age is associated with a decrease in gastric many musculoskeletal tissues with aging) may contrib- acidity, which affects the absorption of calcium and vita- ute to the observed age-related decrease in water con- min B12. When prescribing calcium supplementation for tent. Collagen bundles in these tissues are highly ori- elderly patients, it is important to remember that a high ented, a property that contributes to the mechanical percentage of them will be naturally achlorhydric or characteristics of the tissues. Animal studies have sug- taking an H2 blocker and are unable to absorb calcium gested that changes in the collagen organization, includ- carbonate supplements. Such patients must take calcium ing fibril alignment and fiber bundle orientation, occur citrate supplements, which can be absorbed in the ab- with aging. This change in the organization of the col- sence of stomach acid. A dietary goal for calcium intake lagen fibers may account for the altered mechanics and should be 1,500 mg daily for older individuals, which changes in the ultimate load-to-failure. One study of hu- should be taken in divided doses of 500 mg or less to man ACLs showed decreasing tensile stiffness and ulti- optimize absorption from each dose. mate load-to-failure with increasing age. Another study showed that the ligament-bone complexes in people Although changes occur in the liver that affect the ef- older than 60 years tolerated only one third the load be- ficiency of drug metabolism, these changes do not seem fore failing compared with those of young adults. A re- to be the primary cause of vitamin D deficiency in eld- cent study of human posterior cruciate ligaments found erly people. Serum 25-hydroxyvitamin D levels decline a decrease in collagen fiber diameter and an increase in with age primarily because of decreased sun exposure collagen fibril concentration with aging. and a decrease in the efficiency of vitamin D production in the skin. Changes in the kidney do not seem to affect Nutrition vitamin D levels, except in patients with renal failure. Supplementation with 400 to 800 IU of vitamin D daily Many changes occur to the gastrointestinal system as a is recommended for all older patients. Those who are part of natural aging. Changes in the autonomic nervous vitamin D-deficient (such as patients taking seizure med- system affect colonic motility and can result in constipa- ications, those with sprue, or with irritable bowel syn- tion, whereas other neurologic changes to the anorectal drome) may require a higher dosage. region can result in fecal incontinence. Past diseases also can have an effect on the gastrointestinal system. Eld- Fracture Healing erly patients with a history of diabetes may have special dietary needs to regulate their blood glucose levels and Fracture healing occurs either through a cartilage callus also may have additional bowel motility needs as a re- (endochondral bone formation) when fracture frag- sult of neuropathic changes. Patients may have difficulty ments are not in close apposition, directly onto the sur- swallowing (secondary to stroke) and also may have face of existing bone (appositional bone formation), or special dietary needs after abdominal surgery for ulcers, along a collagen matrix that does not contain any carti- diverticulitis, or colon cancer. All of these concerns can lage (intramembranous bone formation). These modes make it difficult for elderly patients to obtain adequate of fracture healing occur in people of all ages, but the nutrition from food intake alone. speed and efficacy of bone healing declines with in- creasing age. When skeletal maturity is reached, the pe- Protein is essential for the maintenance of muscle riosteum gradually thins and the chondrogenic and os- mass and for the formation of some of the extracellular teogenic potential of its mesenchymal cells declines with components of bone, especially collagen. Follow-up on age. A recent study with rabbits showed that the per- age-related bone loss for patients in the Framingham centage of proliferating mesenchymal chondrocyte pre- study showed lower levels of bone loss if adequate pro- cursor cells did not change between young and old ani- tein was present in the diet. It is widely recognized that mals. However, the same study showed that the absolute patients with low serum albumin levels have difficulty number of proliferating cells decreased in the older rab- with wound healing. Survival rates in elderly patients bits because the size and total cell number in the perios- with hip fractures also is well correlated with adequate teum was decreased. Conflicting evidence has been serum albumin levels. Many elderly patients do not re- found concerning the change in the number and the ceive adequate protein in their diets because of dietary proliferative capacity of osteoprogenitor cells. One choices or social or economic concerns; such patients study of human bone marrow that was aspirated from should be encouraged to increase protein intake until the iliac crest showed an age-related gradual decline in serum albumin levels are within normal limits. Some pa- precursor colony-forming units. Another study of hu- tients also have inadequate caloric intake. If patients are man iliac crest bone marrow also found no change in unable to consume adequate calories with their normal precursor cell number or proliferative capacity with ad- vancing age or with the presence of osteoporosis. American Academy of Orthopaedic Surgeons 73

Physiology of Aging Orthopaedic Knowledge Update 8 Delayed fracture healing has been reported both in cognitive impairments, and the side effects of medica- humans and in animals with aging. Because most experi- tions. The risk of falling increases with an increased ments done on fracture healing have involved young or number of impairments, ranging from 8% in patients young adult animals, few data exist on fracture healing with no impairments to 78% in patients with four or in older animals. Experiments performed to assess the more impairments. Abnormal body sway and gait are effect of aging on fracture healing showed no differ- correlated with falls; several tests can be used to assess ences in the biochemical parameters of fracture healing. these risks in patients. The average osteoporotic 80-year- One recent study comparing 6-week-old rats with old patient should be able to perform a single leg stance 1-year-old rats showed no differences in the messenger for at least 12 seconds. An inability to perform this ma- RNA (mRNA) expression of several cytokines and pro- neuver implies compromised balance. Timed get up and teins involved in fracture healing. Despite this similarity go tests (which evaluate a patient’s ability to rise from a in mRNA expression, femur fracture healing was de- chair without using their arms and to transition to be- layed in the older animals. Expression of Indian hedge- ginning gait) and evaluation of heel-toe straight line hog and bone morphogenetic protein-2 was lower in the walking can be used to identify patients at risk for falls. older animals at the time of fracture callous formation and may have contributed to the delay in healing. This Several strategies can be used to prevent falls and same study found differences in the baseline expression injuries associated with falls. Identification of a patient’s mRNA levels in the young rats even over the short pe- risk factors for falling are an important initial step in riod of the study, suggesting that the age and metabolic choosing preventive strategies. Patients with gait abnor- status of the animal or patient must be taken into ac- malities and muscle weakness may require an assistive count when interventions are considered to enhance device such as a cane or walker to improve mobility. The bone repair. In another rat study, fractures in young ani- removal of throw rugs, the installation of safety rails in mals were healed in 40 days and the bones had normal the bathroom and on stairs, the use of nightlights, and mechanical properties by 4 weeks. In the older animals, the use of proper shoe wear can help to reduce the risk fracture healing was delayed to 80 days, and normal me- of falling posed by home hazards. Other preventive chanical properties were not regained until 12 weeks. measures may include the withdrawal of medications The delayed return of mechanical properties may reflect that are associated with falls (especially psychotropic the increased incidence of hardware failure in older pa- drugs). Exercise programs such as tai chi and physical tients with fractures. The bone holding the hardware in therapy can improve balance, muscle strength, and flexi- place actually fails in many older patients, rather that bility, reducing falls by 46%. The use of energy- the hardware itself breaking before fracture healing is absorbing flooring and hip protectors are indicated for complete. patients who are at high risk for injury from falls. Hip protectors are made of molded polypropylene or poly- The age-related decline in the rate of fracture heal- ethylene and worn as a part of an undergarment over ing may be explained by several mechanisms. Aging is normal underwear. The use of hip protectors has been related to a general functional decline in the homeo- shown to reduce the average patient’s overall risk of hip static mechanisms of skeletal tissues. There is a decline fracture by 60%, and by more than 80% if the hip pro- in the expression of osteoinductive cytokines and tector was being worn at the time of the fall. Patient growth factors both at baseline and with injury in older compliance in using this device is often less than 60% animals, caused in part by a reduction in the inflamma- because of the bulkiness of the undergarment and the tory response to injury. In addition, the bone inductive difficulties some patients have donning and removing it. potential of demineralized bone matrix decreases with aging. The inductive potential of bone matrix appears to Exercise has been shown to decrease the rate of de- be growth hormone-dependent; growth hormone secre- cline of many physiologic changes associated with aging tion decreases with age. by improving cardiac function, strength, balance, and flexibility. Muscle strength reliably increases in older pa- Exercise and Fall Prevention tients who begin resistance exercises, and improvements in muscle mass and flexibility also are seen. Weight- More than 300,000 hip fractures occur annually in the bearing exercises also decrease bone density losses. An United States, and over 90% of these fractures are asso- exercise program for older patients must take into ac- ciated with a fall. Each year, 30% of community dwell- count their functional limitations and goals, modifying ing people older than 65 years fall, and nursing home the choice of exercises as well as the frequency and du- dwellers average 1.5 falls per person per year. A variety ration of the training to increase compliance with the of factors increase the risk of a fall, including muscle program. Elderly people can improve their functional weakness, abnormalities in gait or balance, poor vision, independence, increase mobility, and reduce the risk of decreased sensation in the lower extremities and associ- falling by participating in an exercise program. Elderly ated changes in proprioception, functional limitations, patients who maintain a reasonable level of exercise tol- erance or who can be rehabilitated to this level of activ- 74 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 6 Physiology of Aging ity with a proper physical conditioning program can re- Martin JA, Lingelhutz AJ, Moussavi-Harami F, Buck- duce the decline in overall function that occurs with walter JA: Effects of oxidative damage and telomerase advancing age. activity on human articular cartilage chondrocyte senes- cence. J Gerontol A Biol Sci Med Sci 2004;59:324-337. Annotated Bibliography Chondrocyte cell lines were grown at 5% oxygen and at Changes in Body Composition and Functional Reserve 21% oxygen and analyzed for population doublings. Those grown at 21% oxygen reached senescence at 40 population Buckwalter JA, Heckman JD, Petrie DP: An AOA criti- doublings, whereas those exposed to 5% oxygen concentration cal issue: Aging of the North American population. New reached senescence at 60 population doublings. These findings challenges for orthopaedics. J Bone Joint Surg Am 2003; show that oxidative stress, like that seen in injury and inflam- 85:748-758. mation, can lead to premature chondrocyte senescence. This finding is also an important consideration in establishing the A comprehensive review of the physiologic changes that conditions necessary during chondrocyte transplantation. occur with aging and the demands these changes create for the health care system is presented. The authors also address the Ulrich-Vinther M, Maloney M, Schwarz EM, et al: Artic- response the health care system may need to make to serve ular cartilage biology. J Am Acad Orthop Surg 2003;11: the aging population of North America. 421-430. Cummings SR, Melton LJ: Epidemiology and outcomes A comprehensive review is presented of the current infor- of osteoporotic fractures. Lancet 2002;359:1761-1767. mation available about the biology of articular cartilage and the effects on this tissue on various injuries and diseases in- The rate of osteoporotic fractures is increasing with the cluding osteoarthritis, osteochondral fracture, and microscopic aging of the world population. The authors note the need for damage. This article also summarizes the cytokines and growth epidemiologic research to identify those individuals most at factors that are involved in both normal tissue metabolism and risk for fracture (especially hip fracture) and for treatment response to injury. A discussion of the rationale and evidence with medication. Financial resources for treatments could then of effectiveness of currently available treatments for osteoar- be allocated to those most at risk for a disabling fracture. thritis precedes a discussion of future strategies for treatment based on new information about articular cartilage biology. Stenderup K, Justesen J, Eriksen EF, et al: Number and proliferative capacity of osteogenic stem cells are main- Age-Related Changes in Intervertebral Disk tained during aging and in patients with osteoporosis. J Bone Miner Res 2001;16:1120-1129. Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF, Nerlich AG: Classification of age-related changes in The results of a study that examined the effect of aging lumbar intervertebral discs: 2002 Volvo Award in basic and osteoporosis on the mesenchymal stem cell population are science. Spine 2002;27:2631-2644. presented. Iliac crest bone marrow from 23 younger patients (age range, 22 to 44 years), 15 older patients (age range, 66 to A thorough analysis of the histologic age-related changes 74 years), and 13 osteoporotic patients (age range, 58 to 83) that occur to the human intervertebral disk is presented. One were cultured in vitro. Total colony-forming units, colony sizes, hundred eighty intervertebral disk specimens from a broad and cell density per colony were assessed. Results showed no age range of individuals (range, fetal to 88 years of age) were differences for any group in the number and proliferative ca- analyzed for histologic changes in the cells, matrix, blood sup- pacity of osteoprogenitor cells. The results of this study were ply, and overall structure. The study is the first to provide his- in contrast with some previous studies and suggest that the de- tologic evidence of the changes in blood supply that occur to fective osteoblast functions seen with aging and osteoporosis the disk and precede the structural and cellular changes that are caused by some other factor. are associated with aging and degeneration of the interverte- bral disk. Age-Related Changes in Articular Cartilage Age-Related Changes in Tendon, Ligament, and Joint Capsule Degroot J, Verzijl N, Wenting-van Wijk MJ, et al: Accu- mulation of advanced glycation end products as a mo- Sargon MF, Doral MN, Atay OA: Age-related changes lecular mechanism for aging as a risk factor in osteoar- in human PCLs: A light and electron microscopic study. thritis. Arthritis Rheum 2004;50:1207-1215. Knee Surg Sports Traumatol Arthrosc 2004;12:280-284. This study examined the role of AGEs in the development This study of 36 specimens from patients of various ages of osteoarthritis. Using a canine ACL transection model, the used light and transmission electron microscopy to analyze ul- affected joints were injected with ribose to mimic the presence trastructural differences. Collagen fibers were most variable in of AGE found in older dogs. The development of osteoarthri- diameter in patients age 10 to 19 years. Aging was associated tis and the degree of collagen damage and proteoglycan re- with a decrease in collagen fiber diameter and an increase in lease was then analyzed. AGEs were associated with more se- collagen fibril concentration with a maximum fibril concentra- vere osteoarthritis. tion occurring in patients age 60 to 69 years. American Academy of Orthopaedic Surgeons 75

Physiology of Aging Orthopaedic Knowledge Update 8 Nutrition ciated with falling included female gender, Caucasian race, the presence of an increased number of chronic diseases, use of Meydani M: The Boyd Orr lecture: Nutrition interven- medications, decreased leg strength, poor balance, slower tions in aging and age-associated disease. Proc Nutr Soc walking rates, and lower muscle mass. The authors recommend 2002;61:165-171. that all elderly patients should be assessed for fall risk. This article presents a review of the physiologic changes Lin JT, van der Meulen MCH, Myers ER, Lane JM: that occur with aging and the concomitant socioeconomic fac- Fractures: Evaluation and clinical implications, in Favus tors that influence the dietary regimen of elderly people. Di- MJ (ed): Primer on the Metabolic Bone Diseases and etary modifications to maximize the nutritional intake and ac- Disorders of Mineral Metabolism. Washington, DC, commodate the normal changes of aging are discussed. American Society for Bone and Mineral Research, 2003, pp 147-151. Fracture Healing This chapter describes the compositional changes that oc- Koval KJ, Meek R, Schmitsch E, Liporace F, Strauss E, cur to bone with aging and how this affects the ability of bone Zuckerman JD: Geriatric trauma: Young ideas. J Bone to absorb loads. Fall prevention strategies and environmental Joint Surg Am 2003;85:1380-1388. interventions to decrease applied loads at the time of a fall are discussed. A review of the special considerations needed when treat- ing trauma injuries in elderly patients is presented. Discussion Marks R, Allegrante JP, MacKenzie CR, Lane JM: Hip includes the physiologic changes to bone and soft tissues that fractures among the elderly: Causes, consequences, and occur with aging, as well as patient factors including medical control. Ageing Res Rev 2003;21:57-93. comorbidities and cognitive status issues that are common in this population. The article reviews the current information on A comprehensive review of the physiologic changes of ag- the timing of surgery, anesthesia considerations, implant ing and the risk factors and treatment options for hip fractures choices, fixation enhancement, and postoperative care for the is presented. Comprehensive references for all of the pertinent elderly population. articles published in the past decade, which define risk factors for hip fracture and physiologic factors related to fracture, are Meyer RA Jr, Meyer MH, Tenholder M, Wondracek S, provided. Wasserman R, Garges P: Gene expression in older rats with delayed union of femoral fractures. J Bone Joint Tinetti ME: Clinical practice: Preventing falls in elderly Surg Am 2003;85:1243-1254. persons. N Engl J Med 2003;348:42-49. The levels of mRNA gene expression during fracture heal- Case examples are presented to address the potential ing are the same in young rats and older rats, even though a causes (use of medication, sensory changes, visual distur- delay in fracture healing is observed in the older rats. Lower bances, and cognitive problems) of falls in the elderly popula- levels of Indian hedgehog and bone morphogenetic protein-2 tion. The author presents recommendations for assessing a pa- were detected in the fracture callous of the older rats, which tient during an office visit for fall risk and offers interventions may contribute to slower fracture repair. that can be used based on the deficits noted. O’Driscoll SW, Saris DB, Ito Y, Fitzimmons JS: The Classic Bibliography chondrogenic potential of periosteum decreases with age. J Orthop Res 2001;19:95-103. Bemben MG: Age-related alterations in muscular en- durance. Sports Med 1998;25:259-269. The chondrogenic potential of the periosteum of rabbits ranging in age from 2 weeks to 2 years decreased with aging. Buckwalter JA: Aging and degeneration of the human The most notable change occurred in the total number of cells intervertebral disc. Spine 1995;20:1307-1314. and the thickness of the periosteum. Although the percentage of proliferating cells did not dramatically change with age, the decrease in the total number of proliferating cells correlated with the decline in the chondrogenic potential of the perios- teum with aging of the rabbits. Exercise and Fall Prevention Galea V: Changes in motor unit estimates with aging. J Clin Neurophysiol 1996;13:253-260. DeRedeneire N, Visser M, Peila R, et al: Is a fall just a fall? Correlates of falling in healthy older persons: The Galloway MT, Jokl P: Aging successfully: The impor- Health, Aging, and Body Composition Study. J Am tance of physical activity in maintaining health and Geriatr Soc 2003;51:841-846. function. J Am Acad Orthop Surg 2000;8:37-44. This study evaluated 3,075 high-functioning adults (70 to Hannan MT, Tucker KL, Sawson-Hughes B, Cupples 79 years of age) who were living independently in the commu- LA, Felson DT, Kiel DP: Effect of dietary protein on nity. Frequency of falls over a 1-year period were studied. Re- sults indicated that 24% of women and 18% of men in this study fell at least once during that 1-year period. Factors asso- 76 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 6 Physiology of Aging bone loss in elderly men and women: The Framingham related and species-related changes. J Bone Joint Surg Osteoporosis Study. J Bone Miner Res 2000;15:2504- Am 1976;58:1074-1082. 2512. Quarto R, Thomas D, Liang CT: Bone progenitor cell Martin PE, Grabiner MD: Aging, exercise, and the pre- deficits and the age-associated decline in bone repair ca- disposition to falling. J Appl Biomech 1999;15:52-55. pacity. Calcif Tissue Int 1995;56:123-129. Melton LJ: Epidemiology of hip fractures: Implications Sherman S: Human aging at the millennium, in Rosen of the exponential increase with age. Bone 1996;18:121S- CJ, Glowacki J, Bilizikian JP (eds): The Aging Skeleton. 125S. New York, NY, Academic Press, 1999, pp 11-18. Nishida S, Endo N, Yamagiwa H, et al: Number of os- Taafe DR, Marcus R: Musculoskeletal health and the teoprogenitor cells in human bone marrow markedly older adult. J Rehabil Res Dev 2000;37:245-254. decreases after skeletal maturation. J Bone Miner Metab 1999;17:171-177. Woo SL-Y, Hollis JM, Adams DJ, et al: Tensile proper- ties of the human femur-anterior cruciate ligament-tibia Noyes FR, Grood ES: The strength of the anterior cruci- complex: The effects of specimen age and orientation. ate ligament in humans and rhesus monkeys: Age- Am J Sports Med 1991;19:217-225. American Academy of Orthopaedic Surgeons 77

7Chapter Managing Patient Complaints, Patient Rights and Safety Sally C. Booker Nena Clark, RN Introduction physicians who are perceived as arrogant or uncaring. An understanding of the many factors that can lead to Equal in importance to the skills and knowledge of the patient dissatisfaction is important because many situa- medical practice of orthopaedics are the skills that an tions can be prevented or addressed early, thus minimiz- orthopaedic surgeon must have or develop to deal effec- ing the degree of frustration. Repeated complaints tively with patients from a social and psychological per- should lead the physician to carefully analyze their ap- spective. Experts confirm what may seem obvious, that proach to patients and make improvements in their patients are unlikely to become involved in litigation practice. with physicians whom they like. Also, the overall success of a physician’s practice is highly linked to their ability A patient sometimes may be dealing with an array to interact with patients on a social level. of emotions (sensitivities, embarrassments, fears, shame, anger, and antagonism) when interacting with the physi- Dealing with the disgruntled patient or family can be cian and staff. In addition, health problems produce psy- difficult. Even the most compassionate and skilled or- chological stress that can bring out traits the patient thopaedic surgeon will encounter unhappy patients. It is would not display in normal situations. Prolonged or important to examine the factors that lead to the prob- painful conditions can lead to frustration and despair lem and develop effective strategies for dealing with the that heighten the patient’s emotional response. disgruntled patient. New guidelines have been devel- oped by patient advocacy groups, hospitals, and the fed- Remember that a patient’s complaint is their percep- eral government regarding patient rights and respon- tion of events and attitudes or, in other words, is the pa- sibilities, patient safety, and patient privacy and tient’s version of the truth. When dealing with an angry confidentiality. patient, the physician should avoid becoming angry or defensive and remember that the patient’s anger may be The Social Side of Patient Care based in fear, loss of control, the feeling of being disre- spected or misunderstood, personal issues unrelated to A successful patient interaction often begins with ap- the medical situation, or an organic or psychiatric disor- pearance. A professional, well-groomed physician will der. foster patient confidence, whereas a disheveled or dis- tracted physician may lead to a sense of distrust. Good Several identifiable factors contribute to patient manners and time-honored etiquette go a long way to- complaints, such as inconsistency in communication, ward making a patient feel comfortable. Most impor- promises that are not kept, lack of sufficient details re- tantly, the physician should always treat a patient with garding the diagnosis and treatment plan, perceived respect, dignity, courtesy, graciousness, and compassion. rudeness, lack of understanding regarding known procedural/surgical complications, the perception that The patient’s interaction with office staff is equally the physician and staff are too busy to be concerned important to the overall success of the medical experi- with the patient’s problem, long wait times, and frustra- ence. Patients look for continuity in communication and tion with the inability to “fix” a painful condition. view office staff as an extension of the physician. There- fore, ensuring that office staff is fostering an environ- Complaint Avoidance ment that is patient-friendly is of paramount impor- tance. Complaint avoidance often begins with a well-run of- fice. Avoiding long waits and working with courteous The Disgruntled Patient or Family staff go a long way toward avoiding unhappy patients. The physician’s behavior is also paramount in setting Physicians with excellent bedside manners will encoun- the stage for a positive patient interaction. The physi- ter unhappy patients, although less often than those American Academy of Orthopaedic Surgeons 81

Managing Patient Complaints, Patient Rights and Safety Orthopaedic Knowledge Update 8 Table 1 | Template for Letter of Response to a Patient umented. Next, the nature of any discussions with the Complaint patient and the outcome of the discussion should be re- corded. Finally, all serious complaints should be formal- Paragraph One ized by sending a written response. The Center for Acknowledge the complaint. Thank the patient for taking the time to Medicare and Medicaid Services specifies as a condition express thoughts or concerns. Assure the patient that a complaint repre- of participation that all complaints be addressed with a sents an opportunity for the physician to grow. Let the patient know that written response to the patient in a timely manner. A his/her concerns have been reviewed and discussed. formal letter provides acknowledgment of the complaint Paragraph Two and furthers documents complaint management. When Address the issue. Explain clinical facts in a language the patient can writing a letter, it is important to address the complaint understand. If appropriate, tell the patient what steps have been taken frankly and courteously. In addition to providing the ba- to avoid such an occurrence in the future. If the issue involves potential sis of good medical practice, accurate and detailed docu- liability, a Risk Manager should be consulted before responding to the mentation is vital for cases that result in litigation. patient. Do not respond to the patient in anger. Paragraph Three To create a good system for complaint management, Thank the patient again. Assure the patient that his or her feedback will all complaints and records regarding their management or has been used to discuss the concern among staff, that the concern should be kept in a secure location, and labeled “quality has given you the opportunity to examine your practices, service delivery, improvement” or “quality assurance”. Surveys may be and others. Tell the patient that it is through patient feedback that your useful tools and provide feedback about issues of con- practice or institution can grow. cern to patients that may never reach the formal com- plaint process. Complaints should be viewed as a pro- cian should be warm and cordial, balancing concern and cess of learning or quality improvement for the interest. Generally, it is best for the physician to sit physician and his/her practice. down and speak to the patient at their level. Which Complaints Require a Written The physician needs to be an excellent listener and if Response? he/she senses a problem, it is generally best to address it immediately. A prompt response and action may solve A formal written complaint requires a written response the problem immediately and prevent the issue from de- to the patient. A formal complaint is one where the pa- teriorating to a frustrating situation. tient explicitly asks that a written report be generated. A formal complaint also exists when there is a dispute How to Manage a Patient Complaint about charges based on the patient’s perception of qual- ity of care. If the patient chooses to file a formal com- When a complaint cannot be avoided or diffused, it is plaint, the patient representative in most health care or- important to know how to manage a formal complaint. ganizations will record the patient’s feelings and The first thing to remember is that all complaints must perceptions and submit that report to the attending phy- be taken seriously. The physician is required by law to sician in charge of the patient’s care, or to the manager respond to all formal complaints in writing. Often addi- of the area where the problem occurred. The physician tional information about the complaint is needed and or his/her designee is then obligated to respond to the may be obtained from the patient or an institutional pa- patient in writing (Table 1). In a teaching hospital, it is tient representative. It is advisable at this time to pro- always preferable for the attending physician to respond vide the patient the phone numbers to the patient rep- to the patient rather than the medical or surgical resi- resentative department and allow the patient to take the dent. initiative to contact the department. The patient should be asked if he/she would like to receive a telephone call Health Insurance Portability and or a visit from the patient representative. It is always Accountability Act of 1996 best to allow the patient to make the decision. The Health Insurance Portability and Accountability When speaking with dissatisfied patients, always lis- Act of 1996 (HIPAA), Public Law 104-191 required the ten attentively to their concern. Avoid the pitfalls of in- Secretary for the US Department of Health and Human terrupting or becoming defensive. Acknowledge the pa- Services to implement Sections 261 through 264 of tient’s concerns, even if they are unreasonable. Provide HIPAA. The purpose of these sections, known as the accurate information to help clarify the situation. Be Administrative Simplification provisions, is to improve frank and honest. Avoid negative comments about other the efficiency and effectiveness of the health care sys- health care providers. tem by mandating national privacy standards to enable electronic exchange of consumers’ health information. As with all medical care, documentation is vital The Standards for Privacy of Individually Identifiable when dealing with an unhappy patient. First, the medi- Health Information or the Privacy Rule requires cov- cal assessment should be accurately performed and doc- 82 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 7 Managing Patient Complaints, Patient Rights and Safety Table 2 | What Privacy Rule Says About Disclosure of Pro- Enforcement of Compliance/Penalties for tected Health Information Noncompliance Uses and Disclo- Compliance with the Privacy Rule standards is volun- sures of Patient tary. The OCR of the Department of Health and Hu- Health Information Situations man Services will perform the enforcement of compli- ance. The OCR will investigate reports of violation and Required disclo- To the individual (or individual’s representative) covered entities will be subjected to progressive disci- sures who is the subject of information plinary actions to demonstrate compliance. The OCR To Department of Health and Human Services may impose civil penalties of $100 for each act of non- during a compliance review, investigation or en- compliance of the standards and up to $25,000 per year forcement action for multiple identical violations. However, for criminal penalties, the Department of Justice will perform the in- Permitted uses and To the individual who is the subject of information vestigation, and may impose a fine depending on the se- disclosures (with- For treatment, payment, and health care operations verity of violation from $50,000 to $250,000 and impris- out an individual’s By simply asking for permission from the individual onment of 1 to 5 years (OCR Privacy Brief, 2003, pp 17- authorization) which then give the individual the opportunity to 18). agree or object Patient Safety Issues Use or disclosure of information incident to an already permitted use or disclosure of information The Joint Commission on Accreditation of Healthcare as long as the covered entity has adopted reason- Organizations (JCAHO) is the major accrediting agency able safeguards and the release of information is for hospitals. In 1996, JCAHO developed a sentinel- limited to the minimum amount of protected event reporting policy that encouraged an accredited health information needed to accomplish the health care organization to voluntarily report sentinel pur pose as required by the Privacy Rule events within 5 days and to submit a root cause analysis Public interest and benefit activities (eg, court within 45 days following discovery. JCAHO expects order, serious threat to health or safety, essential health care organizations to perform root cause analyses government functions) of systems and processes of the organization rather than Limited dataset for the purposes of research, pub- the performance of individuals. JCAHO defines a senti- lic health or health care operation nel event as “an unexpected occurrence involving death or serious physical or psychological injury or the risk for No restriction De-Identified Health Information (removal of indi- which a recurrence would carry a significant chance of a vidually identifiable health information) serious adverse outcome.” Not covered under Employment records held or maintained by a cov- Based on the root cause analyses of sentinel events privacy rule ered entity as an employer reported by accredited organizations, JCAHO has iden- Education and other records defined in the Family tified the need to establish national patient safety goals Educational Rights and Privacy Act (Table 3). JCAHO intends to reevaluate the national patient safety goals every year. The goals are to be an- ered entities (such as health plans, health care clearing- nounced in July of each year and to become effective in houses, and health care providers) and its business asso- January of the following year. In July 2002, JCAHO an- ciates to implement the national standards to protect nounced the first National Patient Safety Goals, and in the security and privacy of all individually identifiable January 1, 2003, all accredited health care organizations health information. The Privacy Rule also requires cov- were required to comply with the established patient ered entities to provide individuals an adequate notice safety goals and its corresponding recommendations. of its privacy practices and a description of their individ- JCAHO will allow accredited organizations to imple- ual rights. Furthermore, covered entities are to make a ment alternatives to specific recommendations as long good faith effort to obtain a written acknowledgment of as the alternative is as effective as the original recom- notice of receipt from the individual (Federal Register, mendation in achieving a specific patient safety goal. 2002, p. 53182.). Accredited organizations may submit alternative ap- proaches by completing a “Request for Review of an Protected health information includes demographic Alternative Approach to a National Patient Safety Goal information such as name, address, birth date, Social Se- Requirement” form available on the JCAHO Website: curity number, medical record number, and account http://www.jcaho.org/accredited+organizations/ numbers that relate to: (1) the individual’s past, present, patient+safety/04+npsg/04_npsg_altform.htm. or future physical or mental health, or (2) the provisions of health care to the individual, or (3) the past, present, or future payment for the provision of health care to the individual. (Office of Civil Rights [OCR] Privacy Brief, 2003, p. 4). Table 2 presents information about uses and disclosures of Protected health information. American Academy of Orthopaedic Surgeons 83

Managing Patient Complaints, Patient Rights and Safety Orthopaedic Knowledge Update 8 Table 3 | 2005 National Patient Safety Goals National Patient Safety Goal Associated Recommendation Suggestions for Compliance Improve the accuracy of patient Use at least two patient identifiers (neither to be the Possible identifiers include: identification patient’s room number) when taking blood samples or Patient’s name administering medications or blood products. Patient’s birth date Patient’s medical record number Patient’s Social Security number Patient’s address Example: Ask patient to state his or her name and verify identification with the patient’s wrist band. Name and identification number on the patient’s wrist band should be verified and compared with the name and identification number on the ordered service. Prior to start of any surgical or invasive procedure, conduct All activities in the operating room should cease in order a final verification process, such as “time-out,” to confirm for all members of the surgical team to participate. The the correct patient, procedure, and site, using active—not surgeon or the circulating nurse must state aloud the passive—communication techniques. patient’s name, type of surgery, and location of surgery, as stated in the patient’s informed consent form. All members of the surgical team should respond orally to affirm that the patient’s name, procedure, and location of surgery are correct. Improve the effectiveness of Implement a process for taking verbal or telephone orders Apply to all verbal or telephone orders (not just for communication among caregivers that require a verification “read-back” of the complete medication orders) including all critical test results that order by the person receiving the order. are reported verbally or by telephone. Read-back requirements apply to all including physicians. The National Coordinating Council for Medication Error Reporting and Prevention has made recommendations for improving the use of verbal orders, some of which are: Verbal orders should be limited to urgent situations Entire verbal orders should be repeated back to the prescriber. Verbal orders should be documented in the patient’s medical record, reviewed, and countersigned by the prescriber as soon as possible Standardize the abbreviations, acronyms, symbols not JCAHO has released a “do not use” list of abbreviations to use. such as: U should be written as “unit” IU should be written as “international unit” Q.D., QOD should be written as “daily” and “every other day” Never write a zero by itself after a decimal point (for example, 1.0 mg). Trailing zero after decimal point can be mistaken as 10 mg if the decimal point is not seen. Always use a zero before a decimal point (for example, 0.5 mg). Lack of leading zero before a decimal dose (for example, 0.5 mg) can be mistaken as 5 mg if the decimal point is not seen. MS, MSO4, MgSO4 should be written as “morphine sulfate” or “magnesium sulfate” The Institute for Safe Medication Practices has also published a list of dangerous abbreviations and is available on the Website at: http://www.ismp.org/MSAarticles/improve.htm 84 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 7 Managing Patient Complaints, Patient Rights and Safety Table 3 CONTINUED | 2005 National Patient Safety Goals National Patient Safety Goal Associated Recommendation Suggestions for Compliance Improve the safety of using Remove concentrated electrolytes (including, but not Applies to all concentrated electrolytes high-alert medications limited to, potassium chloride, potassium phosphate, sodium chloride > 0.9%) from patient care units. Eliminate wrong site, wrong Standardize and limit the number of drug concentrations patient, wrong procedure surgery available in the organization. Improve the safety of using Create and use a preoperative verification process, such Applies to all invasive procedures performed in the infusion pumps as a checklist, to confirm that appropriate documents are operating room or special procedures unit (for example, Improve the effectiveness of available such as medical records and imaging studies. endoscopy unit, interventional radiology) that exposes clinical alarm systems patients to more than minimal risk (exception: Reduce the risk of health venipuncture, peripheral intravenous line placement, care-acquired infections placement of nasogastric tube or Foley catheter). Implement a process to mark the surgical site and involve Marking the site is required for procedures involving right the patient in the marking process or left distinction and multiple structures (fingers, toes) or levels (spinal procedures). The American Academy of Orthopaedic Surgeons has developed a checklist for safety called the “Sign Your Site” initiative that involves patients to watch and confirm as the surgeon marks the surgical site. The checklist is available on the Website: http://www.aaos.org/wordhtml/papers/advistmt/ 1015.htm Ensure free-flow protection on all general-use and patient controlled analgesia intravenous infusion pumps used in the organization. Implement regular preventive maintenance and testing of alarm systems. Assure that alarms are activated with appropriate settings and are sufficiently audible with respect to distances and competing noise within the unit. Comply with current Centers for Disease Control and Each of the CDC guidelines is categorized on the basis of Prevention (CDC) hand hygiene guidelines the strength of evidence supporting the recommendation. All category I must be implemented for accreditation purposes. Examples include: When hands are visibly dirty or contaminated, wash hands with a nonantimicrobial soap and water or an antimicrobial soap and water. If hands are not visibly soiled, use an alcohol-based hand rub for routinely decontaminating hands in all other clinical situations. Decontaminate hands before having direct contact with patients. Decontaminate hands after contact with a patient’s intact skin, nonintact skin, body fluids or excretions, mucous membranes, and wound dressings. Decontaminate hands after removing gloves. Before eating and after using a restroom, wash hands with a non-antimicrobial soap and water or with an antimicrobial soap and water. The full report on the CDC guidelines is available on the Website: http://www.cdc.gov/mmwr/preview/mmwrhtml/ rr5116a1.htm Manage as sentinel events all identified cases of This is already required for any outcomes that resulted in unanticipated death or major permanent loss of function unanticipated death or major permanent loss of associated with a health care-acquired infection. function. American Academy of Orthopaedic Surgeons 85

Managing Patient Complaints, Patient Rights and Safety Orthopaedic Knowledge Update 8 Table 4 | Consumer Bill of Rights General Principles Consumer Rights Information disclosure Consumers have the right to receive accurate and easily understood information about health plan, health care professionals, and health care facilities. Choice of providers and plans Access to emergency services Suggestions for health care organizations to ensure this right: Provide reasonable accommodation to meet the needs of patients with language barrier, physical or mental Participation in treatment decisions Respect and nondiscrimination disability. Confidentiality of health information Health care providers educational preparation (eg, education, board certification, recertification) and Complaints and appeals appropriate experience in performing procedures and services. Performance measures such as consumer satisfaction. Provide complaints and appeals processes. Consumers have the right to a choice of health care providers that is sufficient to ensure access to appropriate high-quality health care. Suggestion for health plans to ensure this right: Provide sufficient numbers and types of providers to encompass all covered services. Consumers have the right to access emergency health care services when and where the need arises. Health plans should provide payment when a consumer presents to an emergency department with acute symptoms of sufficient severity – including severe pain – such than a “prudent layperson” could reasonably expect the absence of medical attention to result in placing that consumer’s health in serious jeopardy, serious impairment to bodily functions, or serious dysfunction of any bodily organ or part. Suggestions to ensure this right: Health plans to educate their members about availability, location, and appropriate use of emergency and other medical services. Emergency department personnel to contact the patient’s primary care provider or health plan as quickly as possible to discuss continuity of care. Consumers have the right and responsibility to fully participate in all decisions related to their health care. Consumers who are unable to fully participate in treatment decisions have the right to be represented by parents, guardians, family members, or other conservators. Suggestions for health care organizations/health care providers to ensure this right: Provide patients with easily understood information and opportunity to decide among treatment options consistent with informed consent. Provide effective communication with health care providers for patients with disabilities. Respect the decisions made by patients and/or representatives consistent with the informed consent process. Consumers have the right to considerate, respectful care from all members of the health care system at all times and under all circumstances. An environment of mutual respect is essential to maintain a quality health care system. Suggestion for health care organizations to ensure this right: Provide health care services to patients consistent with the benefits covered in their policy or as required by law based on race, ethnicity, national origin, religion, sex, age, mental or physical disability, sexual orientation, genetic information, or source of payment. Consumers have the right to communicate with health care providers in confidence and to have the confidentiality of their individually identifiable health care information protected. Consumers also have the right to review and copy their own medical records and request amendments to their records. Suggestion to ensure this right: Compliance with the Privacy Rule standards. Consumers have the right to a fair and efficient process for resolving differences with their health plans, health care providers, and the institutions that serve them, including a rigorous system of internal review and an independent system of external review. Suggestion to ensure this right: Internal and external appeals systems and procedures should be made available to patients and resolution should be performed in a timely manner and/or consistent as required by Medicare (eg, 72 hours). 86 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 7 Managing Patient Complaints, Patient Rights and Safety Table 4 CONTINUED | Consumer Bill of Rights General Principles Consumer Rights Consumer responsibility Some of the consumer’s responsibilities include: Maximizing health habits, such as exercising, not smoking, and eating a healthy diet Become involved in specific health care decisions Work collaboratively with health care providers in developing and carrying out agreed upon treatment plans Disclose relevant information and clearly communicate wants and needs Use the health plan’s internal complaint and appeal processes to address concerns that may arise Avoid knowingly spreading disease Become knowledgeable about his or her health plan coverage and health plan options (when available) including all covered benefits, limitations, and exclusions, rules regarding the use of network providers, coverage and referral rules, appropriate processes to secure additional information, and the process to appeal coverage decisions Show respect for other patients and health workers Make a good-faith effort to meet financial obligations Patient Rights and Safety premise that a physician’s bedside manner and interpersonal skills can help influence patient satisfaction is discussed. The Advisory Commission on Consumer Protection and Quality in Health Care Industry, appointed by President Joint Commission on Accreditation of Health Care Or- Clinton in 1997, drafted a “consumer bill of rights.” ganizations web site: 2004 National Patient Safety Health care organizations, both public and private, have Goals. Available at: http://www.jcaho.org/. Accessed De- adopted the general principles as cited in the consumer cember 18, 2003. bill of rights (Table 4). The JCAHO’s National Patient Safety Goals are listed, Annotated Bibliography which were developed as a result of lessons learned from sen- tinel events reported by health care organizations. Bartlett EE: Physician stress management: A new ap- National Coordinating Council for Medication Error proach to reducing medical errors and liability risk. Reporting and Prevention web site: February 20, 2001. J Health Care Risk Manag 2002;22:3-6. Available at: http://www.nccmerp.org/council/council 2001-02-20.html. Accessed December 18, 2003. This article focuses on the scope and effects of medical stress, conceptual approaches to physician stress control, and Valuable information and recommendations on how to re- stress reduction programs, resources, and research. The author duce medication errors related to labeling and packaging of concludes that stress reduction programs can result in better drugs and other related products, verbal orders and prescrip- patient relations, improved clinical performance, fewer medi- tions, dispensing of drugs, and drug administration are pre- cal errors, and reduced malpractice risk. sented. Danner C: Working with angry patients. Behavioral Sage WM: Putting the patient in safety: Linking patient Medicine Brief, Department of Family Practice and complaints and malpractice risk. JAMA 2002;287:3003- Community Health. Minneapolis, MN, Family Medicine 3005. and Community Health, University of Minnesota, Issue 19, 2001. Customer satisfaction is connected to clinical safety. Health care organizations should document information from This article discusses how anger can influence the patients relevant to patient safety and provide that informa- physician-patient relationship. A three-step model for physi- tion to health care professionals who manage and provide cians for effective physical and mental management of critical health care. and angry patients is outlined. The need for the physician to successfully manage anger in order to avoid a malpractice suit US Department of Health and Human Services web and compromised patient care are also discussed. site: Office for Civil Rights. OCR Privacy Brief: Sum- mary of the HIPAA Privacy Rule. Washington, DC, Hickson GB, Federspiel CF, Pichert JW, Miller CS, GPO, May 2003. Available at: http://www.hhs.gov/ocr/ Gauld-Jaeger J, Bost P: Patient complaints and malprac- privacysummary.pdf. Accessed August 8, 2003. tice risk. JAMA 2002;287:2951-2957. A summary of the HIPAA Privacy Rule such as statutory The topic of this article is physicians who receive a large and regulatory background, who is covered by the Privacy number of malpractice claims, resulting in high costs. The Rule, what information is protected, general principles for us- American Academy of Orthopaedic Surgeons 87

Managing Patient Complaints, Patient Rights and Safety Orthopaedic Knowledge Update 8 age and disclosures, and enforcement and penalties for non- Wang EC: Dealing with the angry patient. Permanante J compliance is presented. 2003;7:77-78. US Department of Health and Human Services web Communication skills for physicians faced with an angry site: Office of the Secretary. 45 CFR Parts 160 and 164 patient are discussed, along with insight on what steps might Standards for Privacy of Individually Identifiable be taken for management of this situation. Health Information; Final Rule. Federal Register. Wash- ington, DC, GPO, August 14, 2002. Available at: http:// Classic Bibliography www.hhs.gov/ocr/hipaa/privrulepd.pdf. Accessed August 8, 2003. Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM: Physician-patient communication: The relationship The entire document of the rules and regulations for the with malpractice claims among primary care physicians Privacy Rule released by the Department of Health and Hu- and surgeons. JAMA 1997;277:553-559. man Services is presented. Rogers C: Communicate, in The American Academy of President’s Advisory Commission on Consumer Protec- Orthopaedic Surgeons, Bulletin. Rosemont, IL, 2003, vol tion and Quality in the Health Care Industry web site: 48, No 6. Consumer Bill of Rights and Responsibilities. Executive Summary. Washington, DC, GPO, November 1997. Available at: http://www.hcqualitycommission.gov/cborr/ exsumm.html. Accessed August 3, 2003. A summary of the Consumer Bill of Rights and Responsi- bilities is presented. 88 American Academy of Orthopaedic Surgeons

8Chapter Selected Ethical Issues in Orthopaedic Surgery Angelique M. Reitsma, MD, MA Kornelis A. Poelstra, MD, PhD Surgical Ethics in the 21st Century come increasingly complex and consent forms get longer, truly informed consent may seem a distant goal. Ethical values that guide surgeons toward concrete ac- tions in the care of patients are provided by ethicists Perceptions of what constitutes physical harm have and by the professional surgical societies. The ethical also changed, albeit in many separate and perhaps less principles of beneficence, nonmalfeasance, respect for obvious ways. Therapies that were once state-of-the-art autonomy, and justice were first presented in 1979 as may now be considered obsolete and perhaps even part of the framework known as principlism. Attention harmful. As new evidence about outcomes became is given in the literature to virtue as an additional moral available, procedures have been reshaped or aban- value to be considered. This renewed interest in virtue doned. Evidence-based surgery continues to expand. It as an ethical requirement is one that is in concert with is no longer sufficient to offer a procedure to a patient surgeons’ perceptions of what it means to be a moral merely based on good intentions without some evidence physician. Professional societies such as the American of benefit. Although modern surgery demands scientific College of Surgeons (ACS) and the American Academy evidence, this requirement is not always so easy to com- of Orthopaedic Surgeons (AAOS) have adopted state- ply with. Although drugs can be tested in a randomized ments referring to an honorable surgeon’s character. controlled trial using a placebo for the control, surgeons The AAOS adopted ten Principles of Medical Ethics must invent more intricate and sometimes more inva- and Professionalism in Orthopaedic Surgery, first in sive placebos. Although challenging this goal is not im- 1991 and most recently revised in 2002. The principles possible, the use of sham surgery is not without contro- are considered standards of conduct, not laws, defining versy. Despite strong criticism of such research designs, the essentials of appropriate behavior. it must be conceded that surgical controls are a difficult issue and surgery itself can possibly have quite a strong Perceptions of a moral physician’s conduct have placebo effect. shifted somewhat during the past century. For example, paternalism was perfectly acceptable in medical and sur- Some predicted developments in the future of sur- gical practice for many years, but since the mid 1960s gery are dazzling and will pose new moral challenges. has been considered a breach of patient autonomy. Ma- One example is the use of computers and robotics to jor improvements in respecting a patient’s autonomy enhance surgical precision and safety. These technolo- were also made in the area of human subject research. gies raise ethical questions about training (Who is quali- Currently, it would be unthinkable to enroll a patient or fied to operate such devices, and how much expertise is a healthy volunteer into a clinical trial without their in- required?), responsibility (Who is responsible for oper- formed consent, a requirement that was not obvious to ating the device, who is legally responsible for break- early investigators. As will be discussed later, informed downs?), and perhaps even fears of too much depen- consent for some types of surgical research is still not dence on machinery (What if it breaks down? Should uniform and is an issue for orthopaedic surgeons to con- the surgeon be skilled to perform the procedure without sider carefully. it? Or will the procedure be abandoned?). In some areas of medicine the patient is the sole Other advances will challenge conventional thinking. medical decision maker, with the physician as mere As discussed in a recent article, some of these develop- counselor, offering a virtual and perhaps overwhelming ments include human cloning, genetic engineering, tis- assortment of available therapies. In surgical situations, sue engineering, limb morphogenesis, intelligent robot- this approach is not always feasible, because some of the ics, nanotechnology, suspended animation, regeneration, decision making occurs while the patient is unconscious and species prolongation. The potential of these new and incapable of making decisions. As procedures be- technologies to disrupt conventional surgical thinking is huge. Of direct impact on orthopaedics in the 21st cen- American Academy of Orthopaedic Surgeons 89

Selected Ethical Issues in Orthopaedic Surgery Orthopaedic Knowledge Update 8 tury are the ethical issues concerning tissue engineering, The patient is extubated 2 days after the index pro- regeneration, and intelligent prostheses. Bioartificial cedures. Severe tissue necrosis becomes apparent at the skin and blood vessel segments already can be found in open wounds on both legs. He is unable to tolerate ele- laboratories. The anticipated long-term goal is the re- vation of the head of the bed more than 30° because of generation of body parts, such as parts of limbs and hip pain. Although he recognized his daughter, the pa- eventually, complete prostheses. Will the orthopaedic tient remains poorly oriented to place and time and co- surgeons of the future simply replace a diseased hip herent conversations with the patient remain very diffi- joint, intervertebral disk, or leg with a new one? To ap- cult. The Mini-Mental Status Examination is consistent proach these new technologies, an ethical framework with early dementia. will be required. Controversy about the management of this case Ethical Approaches to a Patient Dilemma causes moral distress among staff and leads to heated discussions. Therefore, a meeting with the ethics com- Every day, physicians are faced with different dilemmas mittee is requested. Two designees of the committee, a in patient care. These vary from simple issues such as or- lawyer from the Department of Risk Management and dering diagnostic tests or surgical scheduling to more a psychiatrist not involved in the case, are present. Dur- complex issues involving decisions about terminal ill- ing this meeting, the daughter continues to strongly nesses and withdrawal of treatment. Deciding on the refuse consent for further treatment, stating that ‘those morally best course can be challenging. Although it is were his wishes’. Although sympathetic to the surgeons’ impossible to provide formulaic answers to every ethical plight, the attorney states that it would be legally inad- dilemma encountered in clinical practice, ethical theo- visable to continue with the surgical intervention as the ries can guide the orthopaedic surgeon in moral deci- patient’s earlier wishes were conveyed by the power of sion making. An actual case is presented to illustrate the attorney, and it is impossible to clearly establish whether prominent ethical theories and to serve as an example he had any decisional capacity to counter this. A sur- of how to work through the nuances of ethical thinking. geon on the committee thought the attorney’s advice to be ‘outrageous’, stating that not treating this patient The Case amounted to ‘torturing the patient to death’. Mr. B is an 87-year-old man who has severe injuries, but Discussion no head, severe chest, or abdominal injuries, after a head-on motor vehicle collision. He is transferred to a The approaches that dominate the ethical literature are Level 1 trauma center. However, he has several severe deontology and utilitarianism. In addition to these two orthopaedic injuries, including bilateral grade IIIB open major moral philosophies, clinical ethics cannot be un- pilon and segmental tibial shaft fractures, a right closed derstood without taking into account the principles of tibial plateau fracture, a displaced femoral neck frac- biomedical ethics: self-determination, beneficence, non- ture, and a left closed supracondylar femur fracture. He maleficence, and justice. Important ethical principles are also has bilateral comminuted foot fractures and domi- listed and defined in Table 1. nant right arm fractures. Prior to the accident he had lived independently, close to his only daughter. The deontologic approach is also called ‘duty-based ethics’, considering the duties that people have toward Mr. B undergoes emergent irrigation and débride- one another. In this case, it is reminiscent of the familiar ment and A-frame external fixator placement of both notion that the surgeon has certain special duties to- legs. He is then transferred, while intubated and in sta- ward the care of the patient with whom a therapeutic ble condition, to the surgical intensive care unit for fur- relationship has been established. From a strictly deon- ther management. A durable power of attorney had tologic point of view, the right approach would be for been granted to his daughter. She is contacted to obtain the orthopaedists to perform the surgeries they believe consent for urgent right below-knee amputation, repeat are indicated, as it is their duty as Mr. B’s treating physi- irrigation and débridement of the left ankle with possi- cians to give him the necessary treatments. This is what ble amputation, and hemiarthroplasty replacement of the orthopaedic surgeons in this case suggested be done the right hip. Despite repeated requests, she strongly re- and requested consent for, to no avail, leading to the fuses to give approval for any further interventions that ethical controversy in the first place. could prolong her father’s life, despite the understand- ing of the severity of the open fractures and the risk for The utilitarianism approach is the view that actions sepsis or pulmonary complications. She states that she are to be morally evaluated according to the amount of and her father had discussed his wishes many years be- well being they promote. This approach is consistent fore, and he had expressed his desire to die at home with preoccupation with the outcomes or consequences without any nursing home care or dependency on a ven- of an intervention, and is why one treatment is recom- tilator for a prolonged period of time. mended over another if it is more likely to provide a de- sired result. Although this approach has been criticized 90 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 8 Selected Ethical Issues in Orthopaedic Surgery for its “the ends justify the means” image, utilitarianism Table 1 | Principles of Biomedical Ethics must be taken into account in public policy as it strives to emphasize the social consequences of an action. In The four clusters of moral basic principles that guide ethics are: the case of Mr. B, the utilitarian approach may suggest that the intervention is not warranted if the results can- Respect for A norm of respecting the decision-making capacity of not be expected to be favorable. Of course, exactly what autonomy autonomous persons. This principle is expressed by the outcome of an intervention will be is not always the doctrine of informed consent. clear, leading to controversy such as in the present case. Nonmaleficence A norm of avoiding the causation of harm. This Self-determination or respect for autonomy is ex- principle requires intentionally refraining from pressed through an informed consent process. By giving actions that cause harm, and is expressed by: “One an informed consent, the patient chooses the treatment ought not to inflict evil or harm.” that they wish according to his or her own values. If a patient is not capable of making or of communicating Beneficence A group of norms for providing benefits and balancing decisions and if an advance directive is not available, a benefits against risks and costs. This principle is surrogate must make treatment decisions on behalf of expressed by: “One ought to prevent evil or harm; the patient. In this case, Mr. B was considered decision- one ought to remove evil or harm; and one ought to ally incapacitated. His daughter was the indicated surro- do or promote good.” gate decision maker as having been appointed by the durable power of attorney. Even if she had not been ap- Justice A group of norms for distributing benefits, risks, and pointed by the patient, the daughter would have proba- costs fairly. Applicable to both medical care and bly been the obvious surrogate by legal standards as the biomedical research with human subjects. closest next of kin. Although in the absence of written advance directives it is not possible to know without a Two relevant moral philosophical theories that further guide decision doubt what a patient would have wanted, the durable making in biomedical ethics are: power of attorney is designed to convey previously ut- tered (verbal) wishes. In this case, respecting the pa- Utilitarianism A consequence-based moral theory holding that tient’s autonomous decision implies that the request for actions are right or wrong according to the balance withholding further surgery should be honored. of their good and bad consequences. The right act in any circumstance is the one that produces the best If Mr. B’s prior wishes were unknown, his daughter overall result, as determined from an impersonal could have used the best-interest standard to make deci- perspective. sions on his behalf. The best interest for a patient is cal- culated by weighing the benefits against the burdens of Deontology A duty-based moral theory holding that some features treatment. From a legal perspective, it is unlikely that of actions other than or in addition to consequences Mr. B (or his daughter) would be forced to continue make actions right or wrong. with the surgical treatment against his wishes. If the daughter had wanted to continue aggressive limb sal- (Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors. vage treatment despite a limited chance of success it is Contemp Pediatr 2003;20:51-53.) likely that that decision, too, would have been re- spected. Increasingly, however, physicians are question- tion to ensure the patient’s comfort during the following ing their responsibilities to patients in situations where days in the hospital. they consider the intervention to have no medical bene- fit. The Role of the Institutional Review Board in Modern Orthopaedics Beneficence, or the obligation to do good for the pa- tient, is closely related to the traditional Hippocratic ob- According to the Department of Health and Human ligation to avoid harm, or the principle of nonmalefi- Services, “research” is defined as any systematic investi- cence. The discussion of beneficence here will include gation designed to develop and contribute to general nonmaleficence. In this case, a beneficent act might con- knowledge. “Human subject” is defined as a living indi- stitute the amputation of both lower extremities to pre- vidual about whom an investigator obtains either vent pain and possibly reduce the risk for septicemia. (1) data through interaction or intervention (such as surgery), or (2) identifiable private information. Re- Justice is the point of reference for discussion of ac- search involving human subjects is socially important cess to health care and of distribution of (scarce) re- but morally perilous because it can expose subjects to sources as a matter of social policy, and will be left out risks for the advancement of science. Ethically justifi- of the discussion in this case. able research must satisfy several conditions, including (1) a reasonable prospect that the research will generate In the case of Mr. B, it was decided to honor the ver- the knowledge that is sought, (2) the necessity of using bal advance directive given to the daughter. Plans were human subjects, and (3) a favorable balance of potential developed to provide sufficient medication and pallia- benefits over risks to the subjects, and (4) fair selection of subjects. Only after these conditions have been met is it appropriate to ask potential subjects to participate. American Academy of Orthopaedic Surgeons 91

Selected Ethical Issues in Orthopaedic Surgery Orthopaedic Knowledge Update 8 Assessing if a proposed study met all the criteria proving the system of oversight and enforcing the de- above was initially left to the discretion of the partments’ regulations. physician-investigator, but the dual roles of research sci- entist and clinical practitioner pull in different direc- Regardless of the criticism they endure, IRBs are tions and present conflicting obligations and interest, currently the only formal and federal mechanism to endangering the objectivity of the assessment. There- safeguard the interests of human research subjects. As fore, the responsibility to assess the morality of clinical such, they are an important line of protection to safe- research was given to outside reviewers. In its 1989 revi- guard human participants of clinical research. Most sion of the Declaration of Helsinki, the World Medical IRBs have a home page on their institution’s Website, Association sets as the international standard for bio- on which they provide information and submission medical research involving human subjects this require- forms to download and complete. ment: “Each experimental procedure involving human subjects should be clearly formulated in an experimen- Although the regulations for submitting research tal protocol which should be transmitted for consider- proposals to an IRB are fairly straightforward, uncer- ation, comment and guidance to a specially appointed tainties remain as to what protocols should and should committee independent of the investigator and spon- not be reviewed. This is especially the case among clini- sor.” cians who do not regularly perform research with hu- man participants. One common false assumption is that The purpose of the Institutional Review Board retrospective studies do not require IRB review. Retro- (IRB) is to ensure that research with human volunteers spective studies whose results are to be published or is designed to conform to the relevant ethical standards. otherwise professionally shared do need to be reviewed These standards concern protecting the rights and wel- by an IRB. Another common misperception is that a fare of individual research subjects, ensuring voluntary study that is likely to be exempt does not have to be re- informed consent is obtained before participation in a viewed. Officially, only an IRB can make the determina- study, and an evaluation of risks and benefits. Since tion that a study is exempt from full review, although 1978, an additional moral standard has been added by many IRBs will have an expedited review process for the National Commission for the Protection of Human studies that are exempt. The important principle to re- Subjects of Biomedical and Behavioral Research (The member is that all human research should be submitted National Commission). The Commission’s “Belmont for IRB review, whether it entails a simple chart review Report” required IRBs to guarantee equity in the selec- or an extensive intervention. tion and recruitment of human subjects. Another area of debate is the use of innovative or Across the United States, IRBs may have local experimental surgery. The introduction of surgical inno- names, such as “Human Investigations Committee” or vations frequently occurs under the header of therapeu- “Committee for Research Subject Protection,” and have tic intervention, without protocols or consents. This sce- a decidedly local character, although they are required nario has the potential to produce an “informal study” to comply with federal regulations when reviewing ac- without specific study consent, which is currently being tivities involving Food and Drug Administration- examined by the surgical and ethical communities alike. regulated investigational drugs or devices, and when In the future, federal regulations or at least professional reviewing research supported by federal funds. Further- standards will likely be developed to address this area more, all institutions that receive federal research grants of surgery to ensure that new innovations are conducted and contracts are required to file a “statement of assur- with appropriate protection of human subjects while ance” of compliance with federal regulations. In these promoting research and technical advances within the assurances virtually all institutions voluntarily promise field. Presently, the issue of what constitutes a signifi- to apply the principles of federal regulations to all re- cant enough innovation as to require an ethical review search they conduct, regardless of the source of funding. or specific consent is left to the discretion of the sur- geon. The IRB system has long been subject to criticism, accused of stifling creativity and impeding progress. In New Technology in Orthopaedic Surgery many institutions, there is a paucity of surgeons that sit on IRBs, thus risking that decisions about surgical stud- Developing new technologies has been the driving force ies will be made with insufficient surgical knowledge. for improvement in surgery. Innovation is necessary if IRBs have also been accused by bioethicists and patient future gains in patient care are desired. Orthopaedists advocates of providing inadequate protection to pro- are among the most innovative surgeons in the United spective research subjects. Even the US federal govern- States; therefore, attention needs to be paid to the ethics ment criticized the work of several IRBs through its Of- of innovation in orthopaedic surgery. This seems espe- fice of Human Research Protections (OHRP). The cially important given the promise of emerging new OHRP was reconstituted by the Department of Health technologies such as cell therapy or the use of growth and Human Services in 2002 and is responsible for im- factors, for which there currently are no clear federal 92 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 8 Selected Ethical Issues in Orthopaedic Surgery regulations, nor are there any clear guidelines pertaining to patients is greatest during certain definable to innovative surgical procedures. “exposure-prone” procedures. Exposure-prone proce- dures include insertion of a needle tip into a body cav- On a professional level, some attempts have been ity, or the simultaneous presence of a health care work- made to address the issue. The ACS has adopted self- er’s fingers and a needle or other sharp object in a imposed guidelines for emerging surgical technologies highly confined anatomic site. Health care workers who and their application to the care of patients, as formu- are positive for HIV or HBV should not perform lated by the Committee on Emerging Surgical Technolo- exposure-prone procedures unless they have obtained gies and Education. The introduction of new technology expert counsel regarding the circumstances under which to surgeons and the public must be done ethically in ac- they may perform such procedures. Furthermore, health cordance with the ACS Statement on Principles. These care workers should inform patients of their infection Principles require prior and continued IRB (or equiva- status before conducting exposure-prone procedures. lent) review of the protocol, full description of the pro- Arguably, there are some limitations to the CDC recom- cedure, and informed consent of the patient. mendations. First, they do not specify exposure-prone procedures, leaving such determinations to the “expert As major innovators, orthopaedic surgeons should counsel,” allowing for a diversity of opinions and deci- assume an active role in thinking through the issue of sions impacting individual infected surgeons or health new surgical technologies. By doing so, orthopaedists care workers. In addition, there are no recommenda- can provide practical guidance for their colleagues as tions that restrict professional activities of health care well as take on a leadership role within the entire surgi- workers infected with HCV, although five cases of viral cal community. Differing views often exist on how to transmission from health care workers to patient have best introduce surgical innovations in an ethically re- been documented to date. Of these cases of viral trans- sponsible way. Although there is no uniform policy to mission, three involved surgeons; one gynecologist and date, some truths appear valid and of practical use to two cardiac surgeons. Perhaps the most important short- the innovative orthopaedic surgeon. First, the innovator coming is that serostatus disclosure as proposed by the should be familiar with existing rules and regulations CDC does not improve patient safety, while it violates that govern human subject research, and know if and physician privacy. when an innovation is in fact a research activity. This de- termination includes the question of whether IRB sub- In addition to the CDC recommendations, other mission and the patient’s informed consent is necessary. professional societies such as the American Cancer So- When a proposed surgical procedure falls short of being ciety, American Medical Association (AMA), and definite research but yet is experimental in a narrower AAOS have issued guidelines. Unfortunately, most sense, or when expected risks are deemed significant or guidelines offer no, minimal, or fairly nonspecific guid- risks and benefits are largely unknown because of the ance as to what constitutes exposure-prone procedures. very novelty of the procedure, additional precautions Also, the issue of disclosure to patients remains contro- appear prudent and moral. Reasonable precautions are versial. Most organizations do not recognize a need for prior consultation (for example, with peers, surgeon-in- disclosure; some guidelines still favor either postexpo- chief, and department chair or division head) and full sure or preprocedure disclosure and/or identification of disclosure of the experimental nature of the procedure the source of infection. It has been argued that the law to the patient with perhaps a separate consent form. It should not require health care workers to disclose their would also be necessary to carefully monitor the out- HIV status (or HBV or HCV status) because this is come of the new technology, if only to assess if further seen as an invasion of privacy. formal evaluation in a trial would be warranted. Regardless of the ongoing controversy, some rele- Other Selected Ethical Issues vant experiential information is available to the HIV- or in Orthopaedics HCV-infected orthopaedic surgeon. The AAOS pro- vides practical guidance for HIV-, HBV-, and HCV- The HIV-Positive or Hepatitis B- or C-Positive Health infected orthopaedic surgeons to ensure the maximum safety of the patient. Orthopaedic surgeons infected Care Worker with HCV should always follow strict aseptic technique and vigorously adhere to universal precautions. In addi- There still is considerable debate concerning the man- tion, they should seek medical evaluation and treatment agement of health care workers infected with hepatitis to prevent chronic liver disease. Specific recommenda- B virus (HBV), hepatitis C virus (HCV), or human im- tions for the prevention of HCV transmission from in- munodeficiency virus (HIV). In 1991, the Centers for fected health care workers may be developed as more is Disease Control and Prevention (CDC) issued guide- learned about the virus and its associated risks. HIV car- lines for health care workers infected with HBV or HIV. ries a lower risk of transmission than either HBV or These guidelines set restrictions that still apply today. HCV. There are only two known cases of HIV transmis- The CDC’s guidelines were established with the premise that the risk for transmission of HBV and HIV American Academy of Orthopaedic Surgeons 93

Selected Ethical Issues in Orthopaedic Surgery Orthopaedic Knowledge Update 8 sion occurring from an infected health care worker to a cess to medical care. The obligations of society and the patient, and one instance where transmission is sus- medical profession to treat the medically underserved pected. In 1990, a cluster of six patients was infected by are also discussed. These recommendations are based on a dentist in Florida. In 1997, an orthopaedic surgeon in the 1993 AMA Council on Ethical and Judicial Affairs France transmitted HIV to one of his patients during a guideline. total hip joint arthroplasty. A third case where transmis- sion is suspected concerns an instance of HIV transmis- Also noted in the document, it is stated that physi- sion from an infected nurse to a surgical patient in cians should be encouraged to devote some time to the France. provision of care for individuals who have no means of paying. Although current data indicate that the risk of trans- mitting a blood-borne pathogen in a health care setting Caring for the uninsured, however, can have a con- is exceedingly low, some risk is still present. The ortho- siderable impact on the orthopaedic surgeon’s practice. paedic surgeon should therefore be familiar with the es- One challenge surgeons encounter when attending to tablished guidelines of the CDC and the AAOS. It is im- the uninsured or underinsured is how to discuss treat- portant that all recommendations to prevent the ment plans. According to the literature, the economic transmission of blood-borne pathogens are consistently constraints on the available care are beyond the physi- followed. In general, the guidelines for preventing trans- cian’s control, yet they raise specific ethical issues for mission from patients to health care workers also apply the physician. The question is whether the physician is to preventing transmission from health care workers to obligated to disclose all potentially beneficial medical patients. Additionally, health care workers are encour- treatment, even the ones that exceed ordinary standards aged to know their own HBV, HCV, and HIV infection of care, and which probably will be unaffordable or un- statuses. Voluntary and confidential testing of health available. care workers for blood-borne pathogens is recom- mended. Preventing injuries to health care workers and Another more pressing problem is the question of subsequent blood exposure to patients offers the great- how much care an orthopaedic surgeon is required to est level of protection. Health care workers who have offer the uninsured. This has become more of an issue preexisting conditions, such as exudative lesions or since the implementation of the 1986 Emergency Medi- weeping dermatitis, should refrain from direct patient cal Treatment and Active Labor Act (EMTALA). This care until the condition is resolved. The affected health act mandates that emergency departments provide med- care worker should also refrain from handling patient- ical screening examinations to every person seeking care equipment and devices used to perform invasive treatment regardless of their ability to pay or whether it procedures. If a member of a surgical team is injured is the appropriate point of service. As a result of this during a procedure, the instrument responsible should law, many patients have also been seeking and receiving be removed from the surgical field without being reused specialty care for nonemergent problems, although this on the patient until appropriately resterilized. Addition- was arguably not the original intent of this law. This ally, any disposable items that come into contact with a consequence of EMTALA has left hospitals and physi- health care worker’s blood should be removed from the cians facing a crisis of overcrowded emergency depart- surgical field and discarded into an appropriate biohaz- ments and uncompensated care, which in turn threatens ard bag or container. patient access to quality care. Finally, it is advisable for any orthopaedic surgeon Other current economic and political changes also infected with HIV or HCV to seek expert legal counsel have major impact on hospitals and health care provid- in addition to medical treatment. An infected surgeon ers, such as declines in Medicare payments for graduate may encounter problems with malpractice insurability, medical education, lower disproportionate share pay- even while in remission and expected to practice. Also, ments, growth of Medicaid managed care, increased legal charges have been leveled against seropositive sur- costs resulting from new government regulations such as geons by former patients, even in the absence of infec- the Healthcare Insurance Portability and Accountability tion, claiming psychological stress. Act (HIPAA) and Center for Medicare and Medicaid Services rules, increased costs of malpractice insurance, Care of the Uninsured Patient increased costs resulting from limited house staff duty hours and decreased payments for services provided The AAOS Opinion on Ethics and Professionalism: from Medicare and other third party payers. These fac- Care and Treatment of the Medically Underserved, orig- tors combined place a strain on orthopaedic surgeons, inally formulated in 1998 and recently revised, discusses especially those in public and teaching hospitals that the dismal health care insurance situation in the United shoulder a disproportionate share of the care for the un- States, noting that increasing percentages of Americans insured. are uninsured, underinsured, and have inadequate ac- It appears appropriate to adopt a prudent stance to- ward providing services at no cost. Caring for the unin- sured is an intrinsic part of orthopaedic surgery in a na- 94 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 8 Selected Ethical Issues in Orthopaedic Surgery Table 2 | Why Physicians May Conceal a Medical Error Table 3 | How To Approach Patients After a Medical Error The medical profession values perfection Notify professional insurer and seek assistance from those who might Feelings of shame or guilt help with disclosure (for example, attending physician or risk manager) The admission may damage the physician’s professional reputation Possible drop in referrals or an impact on income Disclose promptly what is known about the event; concentrate on what Desire to maintain the trust of the patient’s family happened and the possible consequences Pressure felt from various sources: Take the lead in disclosure; do not wait for the patient to ask Managed care organizations Outline a plan of care to rectify the harm and prevent recurrence Hospital administration Offer to get prompt second opinions where appropriate Malpractice insurers Offer the option of a family meeting and the option of having other Fear of punishment or, in the case of trainees, dismissal Fear of a malpractice lawsuit representatives (for example, lawyers) present Document important discussions (Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors. Offer the option of follow-up meetings and keep appointments Contemp Pediatr 2003;20:51-53.) Be prepared for strong emotions Accept responsibility for outcomes, but avoid attributions of blame Apologies and expressions of sorrow are appropriate tion where a great number of people cannot afford even (Adapted with permission from Selbst SM: The difficult duty of disclosing medical errors. basic health care. Contemp Pediatr 2003;20:51-53.) Disclosure in Medical Mistakes In another study, three case scenarios were pre- sented that varied in the degree of outcome severity to Each year thousands of injuries and deaths in US hospi- 400 patients. The respondents generally indicated that tals result from medical errors. Errors involving medica- they would be more likely to file a lawsuit if the doctor tions have been reported in 4% to 17% of all hospital withheld information about a mistake that subsequently admissions. Although most of the information available surfaced. About 40% said they would stay with the phy- about medical errors pertains to hospitalized patients, sician after open disclosure of a mistake was made; how- errors can occur anywhere in the health care system ever, only 8% said they would continue to see a doctor (operating room, office, clinic, emergency department, who did not disclose a mistake. Only 12% said they or elsewhere). would sue if the physician informed them of a mistake that did not result in permanent aftereffects. However, The AMA Code of Ethics provides important guide- 20% said they would sue if they found out about a mis- lines for professional practice including medical mis- take that the physician tried to cover up. takes. The Joint Commission on the Accreditation of Healthcare Organizations requires health care workers These results underscore the fact that patients gener- to inform patients when they have been harmed by ally appreciate an open, honest relationship with their medical mistakes. However, this may not always occur physicians. In fact, a good doctor-patient relationship is as illustrated by one survey of house officers where they one of the greatest factors that reduces the risk of a law- reported that they told an attending physician about se- suit if a poor outcome occurs. Prompt disclosure follow- rious mistakes only about half the time; errors were ing a medical error will make the physician appear hon- conveyed to patient or family in only 24% of the cases. est in the event of litigation and trial (Table 3). In contrast, nondisclosure can have significant negative le- In one study, simulated case scenarios involving gal implications as the legal statute of limitations may wrong medications were presented to 150 medical stu- be extended if a physician is found to have knowingly dents, house officers, and attending physicians. The re- and intentionally concealed information from a patient. searchers found that as severity of injury increased, will- ingness to admit an error declined. About 95% of the Evaluation of the Risk/Benefits Ratio in Patients With students and physicians said they would admit an error to a patient when the outcome was minimal. However, Difficult-to-Treat Problems only 79% said they would admit an error that resulted in the death of a patient. Another 17% said they would Acute, high-risk surgery involves operations on patients admit the error if they were asked directly about the who often face a significant risk of morbidity and mor- event. tality without surgery, but for whom surgery itself in- volves a significant risk of morbidity and mortality as Although concealing a medical error violates ethical well. If nonsurgical management indeed involves a codes, fear of professional censure or medical malpractice higher risk than surgery (for example, acute compart- lawsuits can pressure health care workers to be less than ment syndrome, unstable ’open book’ pelvis injury), the forthcoming when a mistake has occurred (Table 2). American Academy of Orthopaedic Surgeons 95

Selected Ethical Issues in Orthopaedic Surgery Orthopaedic Knowledge Update 8 choice seems simple. However, most situations are not searcher. Orthopaedic surgeons sometimes serve as con- as straightforward. Informing the patient about surgical sultants to companies whose products they are studying and nonsurgical statistics becomes important so that the or join a company’s advisory board and speakers’ bu- pros and cons of surgery can be carefully weighed. In reau, as well as enter into patent and royalty agree- general, the “reasonable person standard” requires phy- ments, which may further complicate financial relation- sicians to tell patients about a therapy’s likely complica- ships. Arguably, such complex and considerable financial tions, especially if surgery is more likely to result in in- relationships may create some (perceived) dependency jury, disease, or a patient’s death than an alternative and/or loss of impartiality for the orthopaedic surgeon. treatment. The best interest for patients is then calcu- lated by weighing the benefits against the burdens of The appropriate ethical approach to take when treatment. If the treatment burdens outweigh the bene- faced with financial conflict of interest has been in- fit, then the treatment is considered not to be in the pa- tensely debated, with editorialized pronouncements ap- tient’s best interests. Benefits of treatment might include pearing in the nation’s leading medical journals; societ- saving or prolonging life, alleviation of pain or restora- ies and professional associations have addressed the tion of function to an acceptable level or quality of life. issue of financial conflicts of interest in research. Relationships With Industry For the moral orthopaedic surgeon, it is appropriate to be familiar with such professional statements, and to Most orthopaedic surgeons foster active relationships apply them to everyday practice and research as much with industry partners. Representatives of companies as possible. Although financial conflicts of interest can that produce prostheses, instrumentation, and other de- probably never be fully eradicated, they can be dimin- vices are frequent visitors to orthopaedic departments. ished and brought to a morally more desirable level. Their companies regularly serve as sponsors of depart- One way this can be achieved is by meticulously disclos- mental functions and books for residents, and often ing all financial interests to patients and research sub- fund clinical research when done to study their prod- jects. Although disclosure is not a curative measure as ucts. These types of financial relationships occur in ev- advertising financial ties will not break them, it will ery branch of medicine and are no less common in or- lessen the potential for harm. In addition, orthopaedic thopaedic surgery. For orthopaedic surgeons, it is surgeons should remain vigilant and cognizant of their therefore important to be aware of the potential ethical ongoing relationships with industry, and regularly criti- problems these relationships can bring about, as such cally evaluate the extent of and the impact such ties relationships may cause conflicting interests. These con- have upon their practice and research. flicts of interest may seriously compromise the integrity of clinical medicine, and may hamper the protection and Annotated Bibliography safety of human research participants. The troublesome effects of conflicts of interest emerging from the re- General Reference search setting can be far-reaching, as published results often establish a standard that is followed by physicians American Academy of Orthopaedic Surgeons (ed): treating patients worldwide. Guide to the Ethical Practice of Orthopaedic Surgery, ed 4. Rosemont, IL, American Academy of Orthopaedic A conflict of interest is defined here as any financial Surgeons, 2003. arrangement that compromises, has the capacity to com- promise, or has the appearance of compromising trust in The first edition of this booklet was published in 1991. The clinical care and/or clinical research. There are many Guide provides standards of conduct and the essentials of eth- different types of financial relationships that may sur- ical behavior for orthopaedic surgeons. This book can be con- face between industry and orthopaedic surgeons and sidered required reading for the morally conscious ortho- their institutions. For example, academic orthopaedic paedic surgeon and resident. surgeons may be provided grants to study a sponsor’s drug or product. Such grants may be a major source of American Medical Association Council on Ethical and salary support for investigators and their personnel. It is Judicial Affairs (ed): Code of Medical Ethics: Current also often the case that device manufacturers offer in- Opinions With Annotations. Chicago, IL, American vestigators financial incentives for entering patients into Medical Association, 2000-2001. a study. Although these “enrollment fees” are intended to cover the cost of the subject’s participation, they typi- This text is presently the most comprehensive and current cally exceed these costs. Academic orthopaedic surgeons guide available to physicians in the United States. It is regu- may use the excess funds to support personnel, travel to larly revised and updated, and includes the seven basic princi- meetings, or laboratory supplies and equipment, while in ples of medical ethics and more than 180 opinions of the private practice these bonuses often remain with the re- AMA’s “EJA”-Council on a wide spectrum of topics. Beauchamp TL, Childress JF: Principles of Biomedical Ethics, ed 5. New York, NY, Oxford University Press, 2001, pp 319-320. 96 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 8 Selected Ethical Issues in Orthopaedic Surgery This book is one of the most influential and important ba- www.facs.org/fellows_info/statements/stonprin.html#top. sic texts on bioethics, offering the theory of the four guiding Accessed August 25, 2004. principles: respect for autonomy, beneficence, nonmaleficence, and justice. This website discusses codes of professional conduct and relationships between patients and surgeons. Surgical Ethics in the 21st Century Cronin DC II, Millis JM, Siegler M: Transplantation of liver grafts from living donors into adults: Too much, too American Academy of Orthopaedic Surgeons Principles soon. N Engl J Med 2001;344:1633-1637. of Medical Ethics and Professionalism in Orthopaedic Surgery, 2002. Available at: http//www.aaos.org/ In this article, the rapid implementation of an experimen- wordhtml/papers/ethics/prin.htm. Accessed August 25, tal procedure into transplantation surgery is criticized. 2004. Margo C: When is surgery research? Towards an opera- These ten principles are standards of conduct that define tional definition of human research. J Med Ethics 2001; the essential aspects of honorable behavior. 27:40-43. Moseley JB, O’Malley K, Petersen NJ, et al: A con- This article analyzes the vague definition of clinical re- trolled trial of arthroscopic surgery for osteoarthritis of search in surgery and criticizes the wide implementation of so- the knee. N Engl J Med 2002;347:81-88. called informal research. In this groundbreaking article the authors use sham sur- Reitsma AM, Moreno JD: Ethical regulations for inno- gery as a placebo in the randomized trial, concluding that ar- vative surgery: The last frontier? J Am Coll Surg 2002; throscopic knee surgery for this particular indication is no bet- 194:792-801. ter than a placebo. This article discusses the regulatory gap between the pro- Satava RM: Disruptive visions. Surg Endosc 2003;17: tection of human subjects involved in research and those un- 104-107. dergoing experimental surgery. Results of a survey among sur- geons are presented. This visionary article paints a picture of the future of sur- gery, with technologies that seem like science fiction now but Other Selected Ethical Issues in Orthopaedics may very well be available within decades. ABIM Foundation: American Board of Internal Medi- The Role of the Institutional Review Board in Modern cine, ACP-ASIM Foundation, American College of Phy- Orthopaedics sicians, American Society of Internal Medicine, Euro- pean Federation of Internal Medicine: Medical United States Department of Health and Human Ser- professionalism in the new millennium: A physician vices Website. Protecting Personal Health Information charter. Ann Intern Med 2002;136:243-246. in Research: Understanding the HIPAA Privacy Rule. Available at: http://privacyruleandresearch.nih.gov/ This article presents a discussion on professionalism, prin- pr_02.asp. Accessed February 2004. ciples, and responsibilities for physicians. The Department of Health and Human Services issued the American Academy of Orthopaedic Surgeons Website. Standards for Privacy of Individually Identifiable Health In- American Academy of Orthopaedic Surgeons Advisory formation (the Privacy Rule) under the HIPAA of 1996 to Statement: Preventing the Transmission of Bloodborne provide the first comprehensive federal protection for the pri- Pathogens. Available at: http://www.aaos.org/wordhtml/ vacy of personal health information. Many of those who must papers/advistmt/1018.htm. Accessed February, 2001. comply with the Privacy Rule complied by April 14, 2003. Practical guidance for all clinically active orthopaedic sur- Department of Health and Human Services National geons infected with HIV, HBV, or HCV. Institutes of Health Office for Protection from Research Risks Website. Available at: http://ohrp.osophs.dhhs.gov/ American Academy of Orthopaedic Surgeons Opinions humansubjects/guidance/45cfr46.htm. Accessed Febru- on Ethics and Professionalism: Care and Treatment of ary, 2004. the Medically Underserved. May 1998, revised May 2002. Available at: http://www.aaos.org/wordhtml/papers/ The Office of Human Research Protections (formerly the ethics/1210eth.htm. Accessed August 25, 2004. Office for Protection from Research Risks, OPRR), provides online decision charts, assisting prospective investigators in de- The AAOS applies its principles of medical ethics and pro- termining the need for IRB review and/or the subject’s in- fessionalism in orthopaedic surgery to this pressing issue. formed consent for research. New Technology in Orthopaedic Surgery American Academy of Orthopaedic Surgeons Position American College of Surgeons Website. Statements on Statement. Emergency Department On-Call Coverage. Principles. Revised March 2004. Available at: http:// September 2002. Available at: http://www.aaos.org/ American Academy of Orthopaedic Surgeons 97

Selected Ethical Issues in Orthopaedic Surgery Orthopaedic Knowledge Update 8 wordhtml/papers/position/1157.htm. Accessed August Classic Bibliography 25, 2004. Aronheim JC, Moreno JD, Zuckerman C (eds): Ethics A detailed statement explaining the various effects of in Clinical Practice, ed 2. Gaithersburg, MD, Aspen Pub- EMTALA on orthopaedic surgeons and the Academy’s stance lishers, 2000, pp 17-50. towards it is presented. Blendon RJ, DesRoches CM, Brodie M, et al: Patient McCullough LB, Jones JW, Brody BA (eds): Surgical safety: Views of practicing physicians and the public on Ethics. New York, NY, Oxford University Press, 1998. medical errors. N Engl J Med 2002;347:1933-1940. Recommendations for preventing transmission of hu- A scientific study of the subjective impact of medical er- man immunodeficiency virus and hepatitis B virus to rors as viewed by physicians and the public. patients during exposure-prone invasive procedures. MMWR Recomm Rep 1991;40(RR-8):1-9. Meyer FN: Uninsured healthcare is a growing problem: Putting economic pressures on physicians. AAOS Bulle- Recommendations for prevention and control of hepati- tin 2003; August 51(4). tis C virus (HCV) infection and HCV-related chronic disease: Centers for Disease Control and Prevention. This article examines the growing concerns about issues MMWR Recomm Rep 1998;47(RR19):1-39. related to uninsured health care in the United States. Selbst SM: The difficult duty of disclosing medical er- Statements on emerging surgical technologies and the rors. Contemp Pediatr 2003;20:51-53. evaluation of credentials: American College of Sur- geons. Bull Am Coll Surg 1994;79:40-41. This article discusses how to manage medical mistakes and examines the reasons why disclosure is so difficult. Statement on Issues to be Considered Before New Sur- gical Technology is Applied to the Care of Patients: Tereskerz PM: Research accountability and financial Committee on Emerging Surgical Technology and Edu- conflicts of interest in industry sponsored clinical re- cation, American College of Surgeons. Bull Am Coll search: A review. Account Res 2003;10:137-158. Surg 1995;80:46-47. A discussion of financial conflicts of interest, research ac- Sweet MP, Bernat JL: A study of the ethical duty of phy- countability, and other aspects of industry relationships are sicians to disclose errors. J Clin Ethics 1997;8:341-348. discussed. US National Commission for the Protection of Human United Kingdom Department of Health Website. HIV Subjects of Biomedical and Behavioral Research: The infected health care workers: A consultation paper on Belmont Report: Ethical Principles and Guidelines for management and patient notification. Available at: the Protection of Human Subjects of Research. Septem- www.doh.gov.uk/aids.htm. A list of exposure-prone pro- ber 30, 1978. Superintendent of Documents, US Govern- cedures available at: www.doh.gov.uk/pub/docs/aids.pdf. ment Printing Office. Washington, DC, DHEW Publica- Accessed February, 2004. tion No. 78-0013. Although UK policy obviously does not apply to US sur- World Medical Association Recommendations Guiding geons, it is worthwhile to review the detailed list of exposure- Physicians in Biomedical Research Involving Human prone procedures as a reference. Subjects (document 17.1). Helsinki, Finland, June 1964. Wears RL, Wu AW: Dealing with failure: The aftermath Wu AW, Folkman S, McPhee SJ, Lo B: Do house officers of errors and adverse events. Ann Emerg Med 2002;39: learn from their mistakes? JAMA 1991;265:2089-2094. 344-346. This article discusses what happens after a medical mis- take has occurred and how patients and health care workers can be guided through this situation. 98 American Academy of Orthopaedic Surgeons

9Chapter Outcomes Assessment and Evidence- Based Practice Guidelines in Orthopaedic Surgery William A. Abdu, MD, MS Outcomes Instruments in Orthopaedic measures of care, which are not always under the pa- tient’s or the doctor’s control. Surgery An outcomes instrument is the survey tool or instru- Traditional approaches to clinical research and reports ment used to measure these variables (Table 1). These of surgical interventions are commonly associated with measures are not designed or intended to substitute or the measurement of variables that are either easily replace the traditional measures or clinical end points, available retrospectively from the medical record, or but are to be used in parallel with clinical measures. variables of primary interest to the researcher. These bi- ologic, physiologic, and anatomic measures include mor- An outcomes instrument, in order to be useful, tality, strength measurement, range of motion, and ra- should have clinical sensibility, meaning that the ques- diographic findings. These examples of “hard” outcome tionnaire includes relevant content and is appropriate measures or surrogate outcomes have a very weak asso- for both the patient population and the setting in which ciation with the symptoms and functionality of the “end it is to be used. The feasibility of the questionnaire is de- result” of therapeutic interventions and are only of termined in part by its length, degree of respondent bur- modest relevance to patients and society. The perceived den, ease of scoring and analyzing the results, and the attractiveness of hard measures is that they are believed costs of its use. In constructing a questionnaire, it must to be objective (unbiased judgment), have preservability be reliable; the results must be reproducible from one (radiographs) and have dimension (measurement of ra- time to another or between interviewers. The question- diographic findings). However, if the goal of treatment naire must also have internal consistency. The instru- is to reduce pain and improve function, these constructs ment must be validated, meaning that there are correla- should be measured directly rather than assuming that tions in the expected direction and magnitude with a surrogate hard measures will suffice. variety of external measures that are somewhat differ- ent but are expected to have predictable associations. In More recently, the goals of clinical research have addition, the instrument must be responsive, or have the transitioned to the measurement of variables primarily ability to measure and detect small but clinically impor- critical to the patient, or “soft” outcome measures. tant differences between groups, or over time. These patient-reported outcomes of symptoms, physical function, and health may take several forms and include Outcomes instruments may be used for multiple pur- the assessment by disease-specific measures, general poses as outlined in Table 2. The increasing interest in health measures, and satisfaction measures. These so- the use of outcomes measures is relatively new and is a called soft outcomes measures are more reliable and rapidly evolving methodology. Many issues remain con- more consistent than the traditional, hard measures. troversial, raising several questions, such as: Which mea- Outcomes research often refers to the study of a group surements are important? When and specifically how or cohort of patients often with the same diagnosis, and should they be measured? Are the outcomes instru- relates their clinical and health outcomes to the care ments valid? Are there controls? What other factors (bi- they received. ologic, physiologic, environmental) may influence the measurement results? Are generic measures sufficient Outcomes research also includes methods of analysis or are disease-specific measures also important? Over for small (single site) and/or large (often multiple site) half a century ago, Lembcke noted that “the best mea- databases, small-area analysis, structured literature re- sures of quality is not how well or how frequently a views (meta-analysis), prospective clinical trials, decision medical service is given, but how closely the result ap- analysis, and guideline development. The focus of out- proaches the fundamental objectives of prolonging life, comes research is often on patient-centered outcomes of relieving stress, restoring function and preventing dis- care (patient self-report) rather than on the process ability.” The goal of outcomes research and the use of American Academy of Orthopaedic Surgeons 99

Outcomes Assessment and Evidence-Based Practice Guidelines Orthopaedic Knowledge Update 8 Table 1 | Common Outcomes Survey Instruments Table 3 | Key Factors in the Construct, Design, and Use of Outcomes Instruments General health outcomes instruments Short Form-36 or Short Form-12 general health survey Content, Population, Setting, and Purpose Health Related Quality of Life (HRQOL) Which outcomes are to be measured (biologic, physiologic, function, Sickness impact Profile (SIP Nottingham Health Profile (NHP) general health, quality of life, satisfaction)? Million Visual Analog Scale (VAS) Which population will be studied and in what setting (age, gender)? Duke Health Profile What is the purpose of the study (to describe, predict, measure change, measure one point in time or change over time, or impact of an inter- Patient satisfaction measures vention)? Patient Satisfaction Questionnaire (PSQ) Content Validity Client Satisfaction Questionnaire (CSQ) Patient Satisfaction Survey (PSS) What domains (health concepts) and items are included? Are there important omissions or inappropriate inclusions? Disease or condition specific instruments include Face Validity Spine Do these measures make good clinical sense? Is each question phrased in a suitable way? Oswestry Disability Index (ODI) for low back pain Are the response categories appropriate? Roland-Morris Disability Questionnaire (RMDQ) Is there an overall score summarizing across questions, and how is Waddell Disability Index this score calculated? Low Back Pain Outcome Score (LBOS) Is there correlation with other outcomes measures? Clinical Back Pain Questionnaire (CBPQ) Are these measures predictive of future events? Quebec Back Pain Disability Scale (QBPDS) Feasibility Low Back Pain Rating Scale (LBPRS) Is the instrument easy to use and understand? North American Spine Society Lumbar Spine Questionnaire (NASS Is the instrument acceptable to the population and the clinician/ LSQ) Resumption of Activities of Daily Living Scale (RADL) investigator? Upper extremity What is its format (self-administered, telephone, personal interview, Constant Shoulder Function Scoring System Carpal Tunnel Syndrome Evaluation computer, paper/pen)? Upper Extremity Disabilities of the Arm, Shoulder, and Head (DASH) How long does it take to administer and report the results? Shoulder Pain Score Responsiveness Elbow questionnaire Is the instrument able to detect subtle but clinically relevant change Lower extremity over time or with intervention? Western Ontario McMaster Universities Osteoarthritis Index (WOMAC) Harris Hip Score Trauma Short Musculoskeletal Function Assessment Questionnaire (SMAF) Table 2 | Purposes for Outcomes Instruments instruments can be divided into categories and are out- lined in Table 3. Study of populations in cross-sectional studies (at one point in time) to define the characteristics of a particular patient population with a The major areas of interest in measuring patient- specific condition, resource utilization, and baseline characteristics. centered self-response measurements using outcomes Determination of longitudinal impact on health change, function, and instruments include measure of function (the ability to satisfaction in an individual patient’s care for a specific disease pro- perform specific tasks, covering the domains of physical, cess after intervention (such as elective surgery). social, role, and psychological function), general health As measures for prospective clinical trials to determine the effectiveness perception (integrates the various aspects of health as of a particular intervention (for example, surgical versus nonsurgical reflected by the patient’s subjective global rating), qual- treatment of a specific condition). ity of life (general measure of a patient’s overall well- Development of disease-specific evidence-based medicine clinical path- being), and satisfaction (measure of treatment impact, ways and guidelines for unifying the process of delivering effective process of care, results, and quality of life). medical care. Although these factors are all distinct concepts, they outcomes survey instruments is to help define the re- may or may not tract together throughout the patient’s sults of interventions and assess these desired end re- course or treatment. For example, resolution of pain sults. may not correlate with radiographic findings, ability to function, return to work, or satisfaction. Thus, outcomes The construct, design, and use of outcomes instru- should be measured in multiple dimensions. Compres- ments are rapidly evolving. Key factors about outcomes sion of various dimensions of outcome into a single uni- dimensional scale obscures the detection of these possi- ble variations in responses. 100 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines Outcomes measures are generally disease-specific or Figure 1 Basic design of a randomized clinical trial. IRB = Institutional Review Board; general health (generic) surveys. Disease-specific instru- DSMB = Data and Safety Monitoring Board. ments such as the Oswestry Disability Questionnaire for low back pain are designed to measure disease-specific functional status, and have the advantage of capturing disease-specific dysfunction in greater detail. General health or generic surveys such as the Short Form-36 are important for detecting complications or ill effects of treatment that extend beyond a specific disease or con- dition. They also make it possible to compare the impact of treatments of specific disorders with the impact of treatment of other diseases. These surveys help deter- mine the cost effectiveness of various treatments. A core set of survey instruments should include measures of symptoms, functional status, overall well-being, and work disability. Several important factors must be considered for outcomes measures to be a valid indicator of health care quality. First, the process of medical care must actually affect the outcome. Second, the measure of outcome must be valid, reliable, and responsive to changes in a patient’s health status. Third, sufficient information must be collected about comorbid conditions and patient de- mographics when studying the outcomes of a particular condition or treatment. Fourth, patient compliance and the timing of a survey are critical to obtaining represen- tative measures. Finally, if the variable of interest is a rare event (such as death), then the population under study must be large enough to make valid comparisons between treatments or conditions. Design of a Prospective Clinical Trial mary outcome of interest. In a prospective clinical trial, the researcher poses a question, intervenes, and follows The major source of information for clinicians is the the direction of inquiry forward. The events of interest published literature, and almost all knowledge in ortho- occur after the onset of the study (Figure 1). paedics is based on information that has appeared in texts and journals. However, factors affecting the valid- In clinical trials, the ability to determine the better ity of many currently reported clinical studies include of two treatments is the product of the trial hypothesis, the lack of randomization of patients, inadequate study the data elements chosen to evaluate the treatments in design, and missing sample size calculations, therefore question, the magnitude of change in the scores over rendering many studies unable to adequately answer the time needed to consider one treatment preference, and research question. Other factors include the lack of sample size for power of study. All research studies are standardized study definitions and measures, poor de- subject to invalid conclusions because of bias, confound- scriptions of study patients and confounding variables, ers, and chance. Bias is the nonrandom systematic error inadequate and unclear follow-up, and absence of in study design. It is an unintentional outcome of factors patient-centered outcomes measures. Given the absence such as patient selection, performance, and outcome de- of a firm knowledge and research on which to base clin- termination. A confounder is a variable having indepen- ical decisions, it is no wonder that significant practice- dent associations with both the exposure and the out- pattern variation exists. come, and thus potentially distorts their relationship. Common confounders include age, gender, socioeco- In order to be most efficient at designing and col- nomic status, and comorbidities. Chance can lead to in- lecting the appropriate data elements, the clinician must valid conclusions based on the probability of type I er- give careful thought to the exact questions to be ad- ror (concluding there is a difference when none exists dressed and the study design to answer the research equal to the p value) and type II error (concluding that questions. The formulation of an appropriate research there is no difference when one truly exists). Thus, the question informs the clinician and researcher about the appropriate sample size and power calculation must be most appropriate data elements to be collected, the pri- American Academy of Orthopaedic Surgeons 101

Outcomes Assessment and Evidence-Based Practice Guidelines Orthopaedic Knowledge Update 8 Table 4 | Common Formulas for the Determination of Sample Size Study Design and Type of Error Sample Size Formula Studies using paired t test (before and after studies) with alpha (type I) N = (za)2 · (s)2 = total number of subjects error only (d)2 Studies using t test (randomized controlled trials with one experimental N = (za)2 · 2 · (s)2 = number of subjects/ group and one control group, considering alpha error only) (d)2 Studies using t test considering alpha and beta errors N = (za+ zb)2 · 2 · (s)2 = number of subjects/ group (d)2 Tests of difference in proportions considering alpha and beta errors N = (za+ zb)2 · 2 · p(1-p) = number of subjects/ group (d)2 N = sample size, za = value for alpha error (equals 1.96 for P = 0.05 in two-tailed test), zb = value for beta error (equals 0.84 for 20% beta error = 80% power in one-tailed test), (s)2 = variance, p = mean proportion of success, d = smallest clinically important difference to be detected determined before the initiation of the study to avoid power is equal to 1-beta error. Therefore, in calculating chance, given the smallest clinically relevant differences a sample size, if the investigators accept a 20% possibil- in outcome and the variability in measures. It is neces- ity of missing a true finding (beta error = 0.2), then the sary to have sufficient statistical power (the ability to study should have a statistical power of 0.8 or 80%. That detect a difference of the magnitude required if one means that the investigators are 80% confident that truly exists) to determine whether the efficacy of one they will be able to detect a true difference of the size treatment is superior to another. that they specify with the sample size that they deter- mine. By design, the investigator accepts a 20% chance Sample Size Calculations of false-negative results. The best way to incorporate beta error into the study is to include it beforehand in One of the more difficult issues determining the success the determination of sample size, which is simple to do, of a study is sample size calculation, which must be but it increases the sample size considerably. A large done prior to the initiation of the clinical study. Sample sample size can be viewed as a means of decreasing the size calculations are always an approximation, yet they false-negative results rate when a real treatment effect are important in providing information about two study exists. design questions. The first is how many subjects should participate in the intended study and second, is this For the comparison of groups in which the data set study worth doing if only N number of subjects partici- variance is high, when the investigators want the answer pate? In some cases the study may not be worth doing if to be close to the “true” value, and when the smallest the likelihood of demonstrating a significant clinical ef- clinically significant difference detected is to be ex- fect or difference between two treatments is very small tremely small, a large sample size will be required. In or if the population needed to answer the question is calculating the sample size, investigators must first de- extremely large. Variables critical to determining sample termine the appropriate formula to be used, based on size include data type (paired data [observations before the type of study and the type of error to be considered. and after treatment in the same study group] and un- Four common formulas for the determination of sample paired data [observations in an observation and control size are listed in Table 4. As the complexity of the group], consideration of error type (beta errors [type II groups and error types increase, the N correspondingly or false-negative errors] and alpha errors [type I or increases. false-positive errors]), variance in data set (large and small), alpha level (usually set at P = 0.05 in the two- The adverse affects of bias, confounding, and chance tailed test and confidence interval of 95%), beta level can be minimized by proper study design and statistical (usually set at 0.2), and determining the size of the de- analysis. The purpose of prospective studies is to mini- tectable difference between groups. mize bias from selection, information, recall, and nonre- sponder factors. Randomization minimizes selection Beta Error Importance bias and equally distributes the confounders. Random assignment implies that each individual has the same Statistical power refers to the sensitivity for detecting a chance of receiving each of the possible treatments and mean difference in a clinical study, whereas sensitivity that the probability that a given subject will receive a refers to the same difference in a clinical test (statistical particular allocation is independent of the probability power + beta error = 1.0). This means that the statistical that any other subject will receive the same treatment 102 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines assignment. The decision regarding study design and Table 5 | Barriers to Overcome Before Proceeding With data collection may be the most critical part of planning Prospective Randomized Clinical Trials Involving a a clinical trial. The study design is the map from which Surgical Procedure data collection follows, and which enables the physician to accurately produce data forms and collect data. The History prospective randomized clinical trial with concurrent controls is the gold standard in clinical research. Tradi- Competition tional randomized trials tell about the efficacy (whether the treatment can work under ideal circumstances), Equipoise/uncertainty whereas outcomes research more typically studies the effectiveness (how well a treatment works when applied Infrastructure in routine care). Effectiveness is a function of efficacy, but also of diagnostic accuracy, physician skill in apply- Study design and statistics ing treatment, and patient compliance with prescribed regimens. Any of these factors may create a gap be- Surgical trials tween efficacy and effectiveness. The prospective ran- domized clinical trial provides not only qualitative con- Events clusions about whether a treatment is better, but also quantitative estimates of the extent to which it is better. Risk assessment Barriers to a Randomized Clinical Trial Commitment There are many barriers that must be overcome to clinical epidemiology can affect study design and statis- proceed with a prospective randomized clinical trial in- tics. The study design and implementation depends volving a surgical procedure (Table 5). History has not heavily on and will be most successful when a core mul- favored the validation of surgical procedures by ran- tidisciplinary group of researchers (including clinicians, domized trials. Many commonly performed surgical pro- epidemiologists, and statisticians) all contribute their ex- cedures were introduced well before randomized trials pertise to the project. became established in medicine. Once a procedure be- comes accepted, it is difficult to test it against a control. The inherent variability of surgery requires precise For many common procedures performed, orthopaedic definitions of the diagnosis, interventions, and close surgeons cannot fall back on history to explain the lack monitoring of its quality. Surgical learning curves might of rigor in surgical research. cause difficulty in timing and performing randomized clinical trials. Blinding is often difficult in surgical trials, Surgeons can be tempted to ignore evidence that and it is not always possible or necessary. Events are threatens their personal interests because of competi- conditions that may be emergent or life-threatening that tion and prestige. Objectivity about procedures central may cause difficulties with recruitment, consent, and to a surgeon’s reputation is difficult and randomized randomization of clinical trials. Multicenter studies are clinical trials may seem threatening. The literature must often required under these conditions. In addition, com- support and the investigators must feel comfortable parison of surgical and nonsurgical treatments with with a state of equipoise, which is a point of maximum greatly different risks causes difficulties with patient uncertainty, a state of balance or equilibrium between equipoise, and thus recruitment. Finally, all involved two alternative therapies such that there is no prefer- must be absolutely committed to all aspects of the trial ence between treatments. Some also refer to this as the if it is to succeed. uncertainty principle in which clinicians, researchers, and patients acknowledge having hunches about a treat- These barriers to randomized clinical trials have ment’s effectiveness, but that the boundaries (or confi- stimulated some to question the need for randomized dence intervals) around their hunches may run from clinical trials in surgery, and the debate is substantial. one extreme (extremely effective) to the other (harm- For many medical questions, a large amount of evidence ful). If the answer to the clinical research question is has been accumulated through nonrandomized studies. clear, the randomized clinical trial cannot ethically pro- The risk of nonrandomized studies is that the studies ceed with the comparison of a known successful treat- may spuriously overestimate treatment benefits as they ment with a known ineffective treatment. are more susceptible to unaccounted confounding, yield- ing misleading conclusions. Observational studies may The infrastructure within which to perform the study provide sufficient evidence of a procedure’s effective- must be well planned and developed to function effi- ness, but the treatment effect of the procedure must be ciently. This includes knowledge to obtain funding and quite large, and the study well designed to be convinc- to perform statistical analysis. A lack of education in ing. However, it is very difficult to use historical con- trols, obtained under less rigorous scientific standards of data collection, against which to test a new procedure. Many previously well established surgical procedures, once thought to provide significant clinical benefits in the hands of proponents, have been subsequently proven ineffective when tested rigorously in well- American Academy of Orthopaedic Surgeons 103

Outcomes Assessment and Evidence-Based Practice Guidelines Orthopaedic Knowledge Update 8 Table 6 | Steps to be Taken Before Patients Are Enrolled in details of data collection, and the process by which the a Clinical Trial data are collected (paper, telephone, personal interview, computer). The schedule for data collection must be de- Establish eligibility criteria that indicate the presence or required eligibil- fined to determine when patients are to be surveyed ity requirements and the absence of exclusion criteria. throughout the study. To properly select appropriate outcome measures, the investigators must be thoroughly Characterize the demographic and general health of the patients eligible familiar with the disease process under investigation. for enrollment into the clinical trial. This allows for the understanding of associated condi- tions, comorbidities, and other variables that are likely Establish a research study database and include baseline data for as- to influence the clinical results. These other associated sessment for changes in the outcome variable to be measured over the conditions and variables (such as age, gender, body mass course of the clinical trial. index, smoking, education level, medical comorbidities) must be measured to provide for statistical control for Recognize and account for stratification required in the randomization their influence (confounding effects) on the primary and process (such as age, gender, race, diagnosis, smoking, education, secondary outcomes of interest. body mass index, socioeconomic factors). Enrollment in a Clinical Trial Develop a system for the determination of follow-up outcomes surveys Before patients are enrolled in a clinical trial, several and tracking patients throughout the trial duration. steps must be taken and these are outlined in Table 6. The enrollment process is best preceded by pilot testing Assess performance to the protocol and adherence to the clinical trial of the procedures, survey instruments, and the overall research protocol with quality checks of clinical sites and investigators. process on test patients to resolve any potential problems before the initiation of the clinical trial with actual en- Establish a Data and Safety Monitoring Board to independently monitor rolled and randomized patients. When a patient is consid- the progress and safety of the clinical trial. ered for the clinical trial, it is necessary to determine the patient’s eligibility for participation. It is most efficient if designed studies. When comparisons of well-designed all the initial required data are collected at nonrandomized trials of selected medical topics with a single visit, the recruitment/preenrollment/random- very narrow and specific selection criteria for the qual- ization visit, during which baseline data are obtained and ity of the meta-analysis were compared with the results treatment assignment occurs. The initial visit must also of randomized trials for the same diagnosis, very good include a thorough explanation of the clinical trial, ex- correlation was observed between the summary odds ra- plaining the purpose and goals, and obtaining the in- tios. However, the nonrandomized studies tended to formed consent for participation. show larger treatment effects, and the discrepancies be- yond chance were less common when only prospective Once randomization occurs, the patient is a member studies were considered. of a treatment group to which they have been assigned. Postrandomization follow-up visits are then scheduled The impact of the randomized clinical trial can be at predetermined times according to the data collection determined by four factors: the baseline control group’s schedule. Completeness of the follow-up visits is critical risk of an outcome event, the responsiveness of patients to the success of the trial. Incomplete follow-up or with- to experimental treatment, the potency of the experi- drawals may bias the results. To ensure maximum mental treatment, and the completeness with which the follow-up participation, it is critical to minimize the bur- outcome events are ascertained and included. The goal den on the patients by requiring only specific instru- of enrollment must be to enroll those most likely to ments to be completed at each visit. Subjects without have the events that are hoped to be prevented with the telephones or those who plan to move are best ex- experimental treatment. Therefore, to measure the true cluded, and the addition of contact information includ- success of the experimental treatment, these high-risk ing email addresses and names and contact information patients become the group most likely to benefit from of family and friends is critical to maintaining contact. participation in the trial and should be the target for en- This information should be updated at each visit. Pro- rollment. It is also critical that all attempts be made to viding reminders for upcoming visits also ensures com- preserve the sample size and prevent erosion of the par- pliance with longitudinal follow up. Incentives are also ticipants by crossover or dropout in follow up. This al- helpful, whether they are financial or other tokens of lows the detection of events that otherwise might have appreciation for continued participation in the clinical been missed, and it increases the chances of being able trial. to present a “worse-case scenario.” This involves the process by which all experimental patients lost to follow Data collection and maintenance is critical to the up in a positive trial are assigned bad outcomes, and all success of a clinical trial. Data must be secure, safe, and lost controls are assigned a positive outcome. The loss of minimal patients to follow-up enhances the credibility of a trial’s positive conclusions. To ensure proper data collection to assess the impact of an intervention, study investigators must agree on the 104 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines protected. Computerized data must be backed up regu- enough to determine whether or not the treatment has larly. Specific identifying data for each patient is critical been successful, and then to inquire “if not, why not”? for providing and maintaining confidentiality and con- sistency. Dedicated personnel whose responsibilities are Evidence-based medicine is not restricted to ran- specific to data management infrastructure are critical domized clinical trials and meta-analyses. It involves to this process. tracking down the best external evidence with which to answer the clinical questions. Best available external ev- The role of maintaining the safety of the participants idence includes clinically relevant research from both in the clinical trial is performed by the Data and Safety the basic sciences and patient-centered clinical research. Monitoring Board (DSMB). This group is commonly It is when questions are asked about therapy that the composed of five non participating individuals who are nonexperimental approaches should be avoided, as knowledgeable about the disease process and statistical these often lead to falsely positive conclusions about ef- and study design. This committee reviews the summaries ficacy. The randomized clinical trial is the gold standard of safety, accrual and progress of the trial, the quality of for judging whether a treatment does more harm than the data, and blinded interim efficacy and effectiveness good. A systematic review of several randomized trials analyses and reports its findings to the principal investi- provides even greater degree of confidence in the level gator and executive working group. It is also responsible of evidence. for interpreting data on adverse side effects. The DSMB meets every 6 months and makes recommendations to Without current best evidence, a clinical practice the principal investigator and executive working group risks the possibility of rapidly becoming out of date, to regarding actions to ensure patient safety and that par- the detriment of the patient. Three evidence-based med- ticipants are not exposed to undue risks. The mandate of icine strategies have been developed to help the clini- the DSMB should comply with the July 1, 1999 release cian maintain and expand clinically important knowl- of the National Institutes of Health policy for Data and edge: learning evidence-based medicine, seeking and Safety Monitoring, with the primary function being to applying evidence-based medicine summaries generated “ensure the safety of participants and validity and integ- by others, and accepting evidence-based protocols and rity of the data.” The DSMB has direct communications practice guidelines developed by others. with the funding agency and can stop a clinical trial when public health or safety is at risk, or when study Classification of Study Strength goals are not met. Evidence is a critical component in the decision-making Evidence-Based Medicine process for delivering care to patients. For many ques- tions, however, the amount of information is over- Physicians seek to base their decisions on the best avail- whelming and the conclusions often contradictory. Most able evidence. Often this represents experience, teach- surgeons would agree that to make the best decisions, ing, and extrapolations of pathophysiologic principles evidence must be easily available in a comprehensive and logic rather than established facts based on data de- and timely fashion. rived from patients. The advent of randomized con- trolled clinical trials has led to an increase in the quality Levels of evidence, based on the rigor of the study of evidence concerning clinical treatment interventions, design, are a way to sort through and evaluate the liter- making clinical reasoning more comprehensible. The ature. Levels of evidence may also be used in the devel- ability to track down and critically appraise and incor- opment of practice guidelines to weigh the strength of porate evidence into clinical practice has been termed the recommendations, and should be used by practicing “evidence-based medicine.” It is the conscientious, ex- surgeons to wade through the multiple types of evi- plicit, and judicious use of current best evidence in mak- dence to help determine which information is useful to ing decisions about the care of individual patients using them and their patients. patient-centered clinical and basic science research data along with the patient’s values and expectations. Levels of evidence provide only a rough guide to the Evidence-based medicine is not “cookbook” medicine. study quality. Level I evidence may not be available for It is the integration of the best external evidence with all clinical questions, and level II and III evidence can individual clinical expertise and patients choice; neither still have great value to the practicing surgeon. No sin- alone is enough. gle study will provide all the definitive answers. Many of these concepts were proposed by Codman in The classification of levels of evidence was devel- the early 1900s, who implemented the “end result” idea. oped to minimize therapeutic harm to patients by using Codman attempted to put into practice the notion that evidence that is least likely to be wrong. The Center each hospital should follow every patient it treats, long for Evidence-Based Medicine (http://cebm.net/ levels_of_evidence.asp) has created the categories for defining study strength (level of evidence) (Table 7) and grades (Table 8) for recommendations in common use today. Similar levels and grades have also been devel- American Academy of Orthopaedic Surgeons 105

Outcomes Assessment and Evidence-Based Practice Guidelines Orthopaedic Knowledge Update 8 Table 7 | Oxford Centre for Evidence-Based Medicine Levels of Evidence and Journal of Bone and Joint Surgery: Instructions to Authors Therapeutic Studies: Prognostic Studies: Economic and Decision Investigating the Results of Investigating the Outcome of Treatment Disease Diagnostic Studies: Analysis: Developing an Investigating a Diagnostic Test Economic or Decision Model Level I Randomized controlled trial Prospective study Testing of previously developed Clinically sensible costs and Significant difference alternatives; values obtained Level II No significant difference but Systematic review of Level I diagnostic criteria in series of from many studies; multiway sensitivity analyses Level III narrow CIs* studies consecutive patients (with Level IV Systematic review of Level I System review of Level I studies Level V universally applied reference randomized controlled trials Clinically sensible costs and (studies were homogeneous) “gold” standard”) alternatives; values obtained from many studies; multiway Prospective cohort study Systematic review of Level I sensitivity analyses Poor-quality randomized studies System review of Level II studies controlled trial (eg, < 80% follow-up Retrospective study Development of diagnostic Limited alternatives and costs; Systematic review poor estimates Level II studies Study of untreated controls from a criteria based on basis of Nonhomogeneous Level I studies Systematic review of Level III previous randomized controlled consecutive patients (with studies Case-control study Retrospective study trial universally applied reference No sensitivity analyses Systematic review of Level III Systematic review of Level II “gold” standard”) Expert opinion studies studies Systematic review of Level II Case series (no, or historical, control group) studies Expert opinion Study of nonconsecutive patients (no consistently applied reference “gold” standard”) Systematic review of Level III studies Case series Case-control study Poor reference standard Expert opinion Expert opinion *CI = Confidence interval Table 8 | Oxford Centre for Evidence-Based Medicine toms, and health-related quality of life as relevant out- Grades of Recommendation comes of clinical care. The growing efforts toward understanding the clinical study design and the imple- A Consistent Level 1 studies mentation of multicenter prospective randomized clini- B Consistent Level 2 or 3 studies or extrapolations from Level 1 cal trials will provide the best evidence with which or- thopaedic surgeons can develop practice guidelines, studies decrease practice variability, provide evidence-based C Level 4 studies or extrapolations from Level 2 or 3 studies quality care, and allow patients to make informed D Level 5 evidence or troublingly inconsistent or inconclusive stud- choices. The current mandate for quality, necessitating the measurements of results, provides an opportunity to ies of any level develop a more focused approach to the delivery of care by orthopaedic surgeons and a means to document the oped by this group for evidence and actions in the do- quality of the end result. mains: prognosis, etiology/harm, and economic analysis. The Future of Evidence-Based Approaches in Annotated Bibliography Orthopaedic Surgery Centre for Evidence-Based Medicine: Oxford-Centre for Evidence Based Medicine Web site. Levels of evi- Efforts in improving the quality of care delivered to pa- dence and grades of recommendation. Available at: tients remains a continuous work in progress. The rou- http://www.cebm.net/levels_of_evidence.asp. Accessed tine use of validated and standardized outcomes instru- September 1, 2003. ments, not primarily for research purposes but also for use in routine clinical practice, will provide data with This is a comprehensive Web-based resource on evidence- which to assess interventions. It seems likely that there based medicine, including definitions and strategies for search- will be a growing and continued emphasis on evidence- ing for literature on evidence-based medicine. based approaches in assessing functional status, symp- 106 American Academy of Orthopaedic Surgeons

Orthopaedic Knowledge Update 8 Chapter 9 Outcomes Assessment and Evidence-Based Practice Guidelines Ioannidis JP, Haidich A, Papa M, et al: Comparison of Daum WJ, Brinker MR, Nash DB: Quality and outcome evidence of treatment effects in randomized and non- determination in health care and orthopaedics: evolu- randomized studies. JAMA 2001;286:821-830. tion and current structure. J Am Acad Orthop Surg 2000;8:133-139. In this review of 45 topics in selected medical topics, anal- ysis of the results demonstrates larger treatment effects and Deyo RA, Weinstein JN: Outcomes research for spinal discrepancy beyond chance nonrandomized studies compared disorders, in Herkowitz H, Garfin S, Balderston R, Eis- with randomized trials. mont F, Bell G, Wiesel S (eds): Rothman-Simeone: The Spine, ed 4. Philadelphia, PA, WB Saunders, 1999. Kocher MS, Zurakowski D: Clinical epidemiology and biostatistics: An orthopaedic primer. J Bone Joint Surg Gartland JJ: Orthopaedic clinical research: Deficiencies Am 2001; Orthopaedic Journal Club: 1-12. in experimental design and determinations of outcome. J Bone Joint Surg Am 1988;70:1357-1364. This is a primer of clinical epidemiology and biostatistics for the orthopaedic surgeon. Hennekens C: Buring J: Interventional studies, in May- rent S (ed): Epidemiology in Medicine. Boston, MA, Lit- McCulloch P, Taylor I, Sasako M, Lovett B, Griffin D: tle, Brown and Company, 1987, pp178-212. Randomised trials in surgery: Problems and possible so- lutions. BMJ 2002;324:1448-1451. Jekel JJ, Elmore JG, Katz DL (eds): Epidemiology, Bio- statistics and Preventive Medicine. Philadelphia, PA, WB The quality and quantity of randomized trials in surgery is Saunders, 1996, pp 160-171. acknowledged to be limited. Some aspects of surgical trials present special difficulties for randomized trials and these dif- Kaska SC, Weinstein JN: Historical perspective: Ernest ficulties and proposed solutions for improving the standards of Amory Codman, 1869-1940: A pioneer of evidence- clinical trials in surgery are discussed. based medicine. The end result of an idea. Spine 1998; 23:629-633. Sackett DL: Why randomized controlled trials fail but needn’t: 2. Failure to employ physiological statistics, or Keller RB: Outcomes research in orthopaedics. J Am the only formula a clinician-trialist is ever likely to need Acad Orthop Surg 1993;1:122-129. (or understand!). CMAJ 2001;165:1226-1237. Kopec JA: Measuring functional outcomes in persons Causes of failure in randomized controlled trials is neither with back pain: A review of back-specific question- with the teacher nor the trialist, but lies in the mismatch be- naires. Spine 2000;25:3110-3114. tween what is judged necessary to be learned about biostatis- tics and who learns it. Pitfalls in conducting clinical trials are Matthews DE, Farewell VT: Using and Understanding discussed, along with appropriate sample size and eligibility Medical Statistics. Basel, Germany, Karger, 1985, pp 184- and erosion of study participants. 195. Watters WC III: Evidence-based practice and the use of Rosenberg WM, Sackett DL: On the need for evidence- practice guidelines. AAOS Bulletin, August 2003, pp 15- based medicine. Therapie 1996;51:212-217. 16. Sackett DL: Evidence-based medicine. Spine 1998;23: Scientific journals, textbooks, and presentations provide 1085-1086. the cornerstone of good clinical decision making and patient care. The benefits and use of evidence-based medicine to pro- Sackett DL: Why randomized controlled trials fail but duce and practice guidelines in the context of clinical quality needn’t: 1. Failure to gain “coal-face’ commitment and improvement cycle is demonstrated. to use the uncertainty principle. CMAJ 2000;162:1311- 1314. Wright J: Summarizing the evidence. AAOSBulletin, June 2003, pp 51-52. Sackett DL, Rosenberg WM, Muir Gray JA, Hayes RB, Richardson WS: Evidence based medicine: What it is Evidence-based medicine, levels of evidence, and strate- and what it isn’t. BMJ 1996;312:71-72. gies for summarizing evidence are reviewed for the practicing orthopaedic surgeon faced with determining treatment recom- Ware JE, Snow KK, Kosinski M, Gandek B: SF-36 mendations. Health Survey: Manual and Interpretation Guide. Bos- ton, MA, The Health Institute, New England Medical Classic Bibliography Center, 1997. Bombardier C: Outcome assessments in the evaluation Weinstein JN, Deyo RA: Clinical research: issues in data of treatment of spinal disorders. Spine 2000;25:3097- collection. Spine 2000;25:3104-3109. 3099. American Academy of Orthopaedic Surgeons 107

Chapter 10 Clinical Epidemiology: An Introduction Karel G.M. Moons, MSc, PhD Diederick E. Grobbee, MD, PhD Introduction improve the course of the patient with this illness and manifestations? The concept of evidence-based medicine implies that the medical care of individual patients should be based The first question asks for the most probable diag- on results obtained from patient-oriented quantitative nosis or illness given the clinical (such as symptoms, research, rather than on qualitative research or clinical signs, test results) and nonclinical (such as age, sex, and experience. Patient-oriented quantitative research is socioeconomic status) profile of the patient. The answer also referred to as clinical epidemiologic research. to the second question, which asks for the cause of the illness, is often impossible to give. Often, adequate care Traditionally, epidemiologic research focused on the can be provided without knowing the cause of an illness. occurrence of infectious diseases and tried to unravel For example, a torn meniscus can effectively be treated determinants of infectious disease epidemics across pop- by the orthopaedic surgeon without knowing how the ulations. Over time, however, it has been shown that injury occurred. The most challenging question concerns methods used for this type of epidemiologic research a prediction of the clinical course of the patient given (population epidemiology) can be applied in a similar the clinical and nonclinical profile, the underlying ill- manner to investigate clinical questions. Clinical epide- ness, and the possible etiology of that illness (question miology is a term commonly used for epidemiology 3). This prediction of the patient’s clinical course may be dealing with questions relevant to medical practice. Ac- made without considering treatment (baseline progno- cordingly, evidence-based medicine is particularly sis) as well as with the consideration of the effects of served by results from clinical epidemiologic research. possible treatments. Differences in the estimated course The distinction between clinical epidemiology and pop- with and without treatment may guide the physician’s ulation epidemiology may be somewhat artificial be- decision on whether to initiate treatment and in the cause both types of studies use largely the same meth- choice between various treatment options. For example, ods for design and analysis. To properly serve clinical if in a patient with a torn meniscus the probability of an practice and provide for evidence-based medicine, clini- unfavorable outcome (such as pain with activity) is cal epidemiologic studies should address relevant clini- much higher without surgery (baseline prognosis) than cal questions, be validly executed, and should yield re- with surgery, it is likely that surgery would be chosen as sults with sufficient precision. the proper course of action (question 4). Clinical Epidemiologic Research Questions For a physician to meet the challenges of clinical de- cision making, quantitative or probabilistic knowledge is Understanding the challenges related to clinical practice required. Diagnostic research yields quantitative knowl- is essential for understanding the objectives of clinical edge about the probability of disease presence or ab- epidemiologic research. Beginning with the patient- sence given the patient’s clinical and nonclinical profile physician encounter, four principal challenges or ques- (addressing question 1). Results from etiologic research tions in clinical practice can be defined. These questions help the physician to answer questions about the proba- relate back to Hippocrates and form the basis of the bility that a particular disease will occur when a particu- evidenced-based medicine movement. In hierarchical lar risk factor is present (question 2). Prognostic re- order these questions are: (1) What is the illness of the search provides quantitative knowledge about the patient given the clinical and nonclinical profile? probability that a particular outcome (death, complica- (2) Why did this illness occur in this patient at this time? tion, recurrence, or improved quality of life) may occur (3) What will be the course of the patient with this ill- over a particular period of time in relation to the pa- ness and manifestations? (4) How can I, as a physician, tient’s profile, illness, and etiology (question 3). Treat- ment decisions require knowledge about the probability American Academy of Orthopaedic Surgeons 109