Vol 9, No 6, November/December 2001 355 Mobile-bearing knee arthroplasty (MBKA) was introduced in the late 1970s for two main purposes. 1,2 The first was to increase contact area so as to reduce long-term wear, and the second was to recreate normal knee kinematics. A theoretically attrac- tive concept compared with fixed- bearing knee arthroplasty, MBKA was described as an evolutionary advance in total knee design. 3 All MBKA devices can be characterized as involving a moving polyethylene bearing separating the femoral condyle from the tibial tray. Most orthopaedic total joint manufactur- ers have developed or are currently developing an MBKA prosthesis. Theoretically, the increased con- gruity and decreased axial con- straint of an MBKA prosthesis should lead to less penetrative wear of the polyethylene and reduced loosening torque at the prosthesis- bone interface. 4 Other stated bene- fits are improved patellofemoral and tibiofemoral biomechanics with increased maximal flexion. Al- though the reported results with current total knee arthroplasty (TKA) designs have been excellent, 5-7 most surgeons consider youth a rela- tive contraindication. Low-wear MBKA has been suggested as the next evolution in TKA design to ex- pand its indications to include the young, active patient. 8 It is important for the surgeon who is considering using these prostheses to understand the elements of MBKA design, the kinematics of mobile-bearing TKA devices, the constraint-conformity conflict, reduction of wear, the po- tential for bearing dislocation and breakage, stability, clinical results, and indications. Types of MBKA Devices The term MBKA describes a variety of dissimilar knee prostheses that fea- ture a mobile polyethylene bearing that articulates with a metallic fem- oral condyle and a metallic tibial tray. Walker and Sathasivam 9 have classi- fied current designs into four types on the basis of bearing mobility: (1) The “internal-external rotation only” design (Fig. 1) allows the knee to locate to a preferred rotational ori- Dr. Vertullo is Fellow, Division of Orthopaedic Surgery, Duke University Medical Center, Durham, NC. Dr. Easley is Assistant Profes- sor of Orthopaedic Surgery, Duke University Medical Center, Durham. Dr. Scott is Direc- tor, Insall Scott Kelly Institute for Orthopae- dics and Sports Medicine, New York, NY. Dr. Insall is Director, Insall Scott Kelly Institute for Orthopaedics and Sports Medicine. One or more of the authors or the departments with which they are affiliated have received something of value from a commercial or other party related directly or indirectly to the sub- ject of this article. Reprint requests: Dr. Easley, Box 2950, Division of Orthopaedics, Duke University Medical Center, Durham, NC 27710. Copyright 2001 by the American Academy of Orthopaedic Surgeons. Abstract Mobile-bearing knee arthroplasty (MBKA) has potential advantages compared with conventional fixed-bearing total knee arthroplasty (TKA). By allowing unconstrained axial rotation, MBKA can offer greater articular conformity with- out an increased probability of loosening due to increased axial torque. Increased articular conformity minimizes polyethylene contact stresses, thereby reducing linear wear and subsurface fatigue failure. Axial rotation of the platform also enables self-correction of tibial component malrotation. Despite these advan- tages, the long-term clinical results obtained with current MBKA devices are similar to those obtained with well-designed fixed-bearing TKA prostheses, with no data suggesting their superiority. The disadvantages of MBKA include bear- ing dislocation and breakage, soft-tissue impingement, a steep technique learning curve, and concerns about volumetric wear. Hypothetically, longer-term follow- up of MBKA results may reveal a significant difference from fixed-bearing TKA results as the fatigue failure threshold of incongruent polyethylene is exceeded. J Am Acad Orthop Surg 2001;9:355-364 Mobile Bearings in Primary Knee Arthroplasty Christopher J. Vertullo, MBBS, FRACS(Ortho), Mark E. Easley, MD, W. Norman Scott, MD, and John N. Insall, MD Perspectives on Modern Orthopaedics entation. Backward motion of one condyle is accompanied by forward movement of the other. 10 (2) The “internal-external rota- tion about a medial axis” design (Fig. 2) better simulates anatomic motion as the normal knee rotates through a longitudinal axis on the medial tibial plateau. 11,12 (3) The third type allows internal- external rotation and anteroposterior (AP) translation so that the knee can locate at a preferred rotational and translational orientation (Fig. 3). This design relies on ligamentous structures for stability and kinemat- ics. This type can be posterior cru- ciate ligament (PCL)–retaining or PCL-sacrificing and includes menis- cal-bearing designs. 10 (4) In the “guided motion” type, internal-external rotation is allowed, and AP translation is guided by intercondylar cams or guide sur- faces in an attempt to reproduce the AP motion of the natural knee (i.e., roll-back with flexion and roll- forward with extension). This type includes the posteriorly stabilized MBKA devices and designs that have an intercondylar saddle- shaped cam. 13 Posteriorly stabilized designs produce roll-back only with flexion. Saddle designs produce roll-back and roll-forward with flex- ion and extension, respectively. The guided roll-back achieved with these designs in high flexion is preferable to that achieved with rotation-only designs; however, it requires a partially conforming femoral-tibial bearing surface. Some prostheses have two separate fully congruent bearing surfaces on each condyle, one for 0 to 8 degrees of flex- ion and one for 8 degrees to maxi- mum flexion. Mobile-bearing designs also dif- fer in the types of constraint mecha- nisms used to prevent bearing dislo- cation. These include a cone-in-cone articulation of the polyethylene insert with a tray recess, longitudi- nal curved sliding tracks, or a tibial tray post that articulates with a poly- ethylene recess. Some designs also include stops to limit excessive AP translation and/or rotation; how- ever, there are concerns about the generation of polyethylene wear particles. Designs incorporating meniscal bearings on curved tracks to allow axial rotation have been criticized for not allowing AP trans- lation without approaching or re- ceding from each other, decreasing the meniscofemoral contact area. 1 An MBKA prosthesis can be par- tially congruent or fully congruent. The contact area differs between designs as a result of variations in the sagittal radius of the femoral condyle. Partially congruent, or gait- congruent, MBKA devices have large contact areas in the first 20 degrees of flexion that decrease with flexion due to a decreasing sagittal radius (Fig. 4). Gait-congruent pros- theses were designed so as to maxi- mize the contact area in the more important low end of the flexion range while decreasing the sagittal radius, to improve flexion range. Fully congruent MBKA devices have a constant sagittal femoral radius, allowing much larger contact areas. However, fully congruent MBKA prostheses have a theoretical limit of 120 degrees of flexion due to poste- rior impingement of the tibial com- ponent. 10 Some MBKA designs are offered as part of a knee arthroplasty system with the ability to change from fixed-bearing to mobile-bearing intraoperatively once the knee disor- der has been better assessed. Other MBKA systems are stand-alone, with only a mobile-bearing arthro- plasty possible (Table 1). Currently, the only devices approved by the US Food and Drug Administration (FDA) for noninvestigational use are the Low Contact Stress Rotating Platform and the Low Contact Stress Meniscal Bearing Knee (DePuy, Warsaw, Ind). Kinematics of MBKA Devices With progressive flexion, the nor- mal knee undergoes posterior dis- placement (roll-back) of the femur on the tibia and internal rotation of the tibia. This passive motion can Mobile Bearings in Primary Knee Arthroplasty Journal of the American Academy of Orthopaedic Surgeons 356 Figure 1 With the “internal-external rota- tion only” design, there is a cone-in-cone constraint mechanism. Figure 2 With the “internal-external rota- tion about a medial axis” design, there is rotation about a longitudinal axis through the medial tibial plateau. be described as femoral rotation about an axis through the femoral epicondyles coupled with tibial rotation about an axis parallel with medial to the long axis of the tib- ia. 11,12 Roll-back occurs primarily as a function of the PCL, optimizing the quadriceps lever arm in flexion and allowing clearance of posterior structures, thereby increasing the range of flexion. The increased quadriceps lever arm is an advan- tage during downhill walking and stair descent. Although one of the original aims of MBKA was to recreate nor- mal knee kinematics, 1,2 this remains theoretical. The current literature on MBKA kinematic behavior sug- gests that no current design closely mimics the normal knee, but rather shows numerous kinematic abnor- malities observed with fixed-bearing designs. 10,14,15 Abnormalities in- clude paradoxical anterior femoral translation, reverse axial rotational patterns, and femoral condylar lift- off. As would be expected, cruciate- sacrificing and posteriorly stabilized rotating platforms show less AP translation in gait and less variabil- ity between individuals than fixed- bearing designs, due to increased sagittal femorotibial conformity. 10 Christopher J. Vertullo, MBBS, FRACS(Ortho), et al Vol 9, No 6, November/December 2001 357 An in vivo fluoroscopic weight- bearing kinematic analysis of fem- oral roll-back in a cruciate-retaining mobile-bearing device suggested that it behaves unpredictably with abnormal kinematic function, simi- lar to other cruciate-retaining fixed- bearing designs. 16 Roll-back oc- curred at up to 90 degrees of flexion, but anterior translation was ob- served with flexion greater than 90 degrees. An in vitro comparison of rotating- bearing and fixed-bearing knees with either cruciate substitution or cruciate retention demonstrated that all four designs underwent some degree of roll-back, but that both of the cruciate-retaining de- signs showed greater roll-back. 14 However, the cruciate-retaining fixed-bearing knee in that study was less conforming than the cruciate- substituting fixed-bearing and rotating-bearing knees, and there- fore allowed greater AP translation. Normal internal tibial rotation with flexion is reduced in both fixed-bearing and rotating-only mobile-bearing total knees. 10 PCL- substituting and PCL-sacrificing rotating-only designs showed tibial internal rotation with deep knee bends. However, in gait the cruciate- sacrificing rotating-only design underwent paradoxical tibial exter- nal rotation with gait. Rotating-platform and meniscal- bearing MBKA designs have similar or less flexion than fixed-bearing designs. 10,15,17-19 The least amount of flexion occurs with the cruciate- sacrificing rotating-platform de- signs, which often exhibit anterior femoral roll-back in deep flexion. An in vitro study comparing the maximal flexion of multidirectional MBKA, rotating MBKA, AP-translating MBKA, and fixed-bearing TKA de- vices showed no differences. 20 All articulating surfaces had identical A B C Figure 3 With this design, the polyethylene may rotate, translate, or rotate and translate on the tibial baseplate. The polyethylene is in a neutral position relative to the baseplate in extension (A), but with flexion, it rotates (B) and translates (C). Figure 4 Flexion may be enhanced by reducing the posterior sagittal radius of the femoral component relative to its radius in extension. geometry, and the PCL was re- tained in all knees. It has been sug- gested that maximal flexion can be obtained with a mobile-bearing de- sign with a decreasing sagittal con- dyle radius that is posteriorly stabi- lized to achieve predictable femoral roll-back. 10 Theoretically, the self-alignment ability of MBKA tibial platforms can improve patellofemoral mechanics. However, currently there is little kinematic data to support this hy- pothesis. Authors have reported minimal anterior knee pain and no patellar subluxation or dislocation in large series, 10,21 but this may be attributable to excellent surgical Mobile Bearings in Primary Knee Arthroplasty Journal of the American Academy of Orthopaedic Surgeons 358 Table 1 Design Features of Various MBKA Devices * Implant Conformity Bearing Constraint Device Other Features Low Contact Stress Gait-congruent Rotating-only Cone in cone PCL-sacrificing Rotating Platform platform (DePuy, Warsaw, Ind) Low Contact Stress Gait-congruent Meniscal bearings Curved tracks PCL-retaining Meniscal Bearing (DePuy) Low Contact Stress Gait-congruent Multidirectional Cone in cone with PCL-retaining, no Antero-posterior Glide platform AP slide rail restraint to AP glide (DePuy) Self Aligning Knee Fully congruent Multidirectional Tibial tray post PCL-retaining (Sulzer, Austin, Tex) platform Medially Biased Fully congruent Multidirectional Tibial tray post, PCL-retaining, fixed- Kinematics platform subluxation stops bearing intraoperative (Zimmer, Warsaw, Ind) options Two Radii Area Contact Fully congruent Rotating-only Cone in cone, Two separate areas of (Biomet, Warsaw, Ind) platform guided-motion articular contact via saddle inner and outer tracks, posteriorly stabilized Genesis II Mobile Bearing Gait-congruent Multidirectional or Tibial tray post PCL-retaining, fixed- Knee (Smith & Nephew, rotating platform bearing intraoperative Memphis, Tenn) options Oxford Bicompartmental Fully congruent Meniscal bearings Relies on ACL ACL and PCL (Biomet, Bridgend, UK) and PCL Rotaglide (Corin, Fully congruent Multidirectional Tibial tray post, PCL-retaining, fixed- Gloucestershire, UK) platform subluxation stops bearing intraoperative options P.F.C. Sigma Rotating Gait-congruent Rotating-only Cone in cone PCL-sacrificing Platform (DePuy) platform P.F.C. Sigma Stabilized Gait-congruent Rotating-only Cone in cone Posteriorly stabilized Rotating Platform (DePuy) platform Interax ISA (Howmedica, Gait-congruent Multidirectional Tibial tray post PCL-retaining Allendale, NJ) platform Scorpio Knee Gait-congruent or Unidirectional or Tibial tray post PCL-retaining/ (Stryker Howmedica, fully congruent multidirectional sacrificing or Allendale, NJ) platform substituting Link Tack (Waldemar Fully congruent Rotating only Curved tracks PCL-retaining Link, Hamburg, Germany) platform * Abbreviations: ACL = anterior cruciate ligament; AP = anteroposterior; PCL = posterior cruciate ligament. technique rather than prosthetic design. Wear, Conformity, and Contact Stress Reducing the generation of poly- ethylene wear particles improves TKA survivorship by decreasing loosening secondary to aseptic osteolysis. 22,23 Knee wear is a com- plex, multifactorial process affected by a wide range of variables, in- cluding polyethylene quality and processing, sterilization techniques, articular kinematics, lubrication, applied load, and articular topogra- phy. 24 Wear occurs at the superior femur–polyethylene surface and the inferior tibia–polyethylene sur- face (backside wear) in both fixed- bearing and mobile-bearing TKA prostheses. 25 The complexity of the in vivo wear process in TKA is demonstrated by the lack of a coherent theoretical model and conflicting in vitro wear data from studies with nonuniform testing conditions. Researchers have sug- gested that it is unlikely that any in vitro test can become a primary standard for wear measurements. 24,26 Two types of wear patterns most commonly occur in TKA: abrasive wear and fatigue failure, which pro- duces pitting and delamination. 23 Fatigue failure commonly pro- duces delamination and surface pit- ting due to subsurface stress that exceeds the polyethylene failure threshold of 9 MPa. 27 Round-on-flat TKA designs with extremely high contact stresses are an example of nonconforming articulations that are subject to accelerated delamina- tion and pitting. 28 By increasing the articular conformity, subsurface stresses can remain below the stress- yield threshold, preventing fatigue failure of polyethylene. 27 Decreasing abrasive wear is a more complex issue, with the mini- mal contact stress value that initi- ates in vivo wear affected by the polyethylene quality, material pro- cessing, sterilization techniques, lubrication, load, and articular kine- matics. 25 In vitro evidence suggests that wear rates are less dependent on contact stress below 6.9 MPa, a level above which wear rates accel- erate substantially. 26 A recent finite-element analysis examined the effects of different conformity ratios and loads on polyethylene stress levels in total knee prostheses. 29 A ratio of 0 rep- resented a flat-on-round design, and a ratio of 0.99 represented a fully conforming design. Polyethylene stresses were more sensitive to changes in conformity than to load changes. Doubling the load from 3,000 to 6,000 N resulted in less stress increase than changing the conformity ratio from 0.99 (fully conforming) to 0.95. The effect of increasing conformity ratio on the reduction in stress was more pro- nounced for ratios above 0.8. The deleterious effect of a load increase for a flat tibial tray was double that for one with full congruity. By increasing articular conformity and decreasing contact stresses to less than 6.9 MPa, the most com- mon modes of TKA wear can be decreased. However, the disadvan- tages of highly conforming fixed- bearing TKA devices are decreased freedom of motion and resultant increased transmission of torque forces to the bone-prosthesis inter- face. 1,10,30 Mobile-bearing knee ar- throplasty has been proposed as a method to overcome this kinematic conflict of low-stress articulations with free rotation. 1,31 Finite-element analysis suggests that MBKA devices generally achieve lower contact stresses and subsur- face stresses than fixed-bearing TKA devices at heel-strike. 3,25,27,32 Predic- tions of low linear wear for highly congruent mobile-bearing knees derived from in vitro analysis have been matched by retrieval studies of the fully congruent mobile-bearing Oxford Knee Replacement (Biomet, Bridgend, UK). 33,34 The unicom- partmental Oxford device had a mean wear of 0.036 mm per year, and the bicompartmental device had a mean wear of 0.043 mm per year, both of which compare favor- ably with wear rates in highly con- forming total hip replacement. Failure due to wear or osteolysis has been reported at very low rates in clinical series of MBKA. Sorrells 35 reported the results with a partially congruent rotating platform at a follow-up interval of 1 to 11 years. The failure rate due to wear was 0.2%; the single failure in that study was ascribed to poor-quality poly- ethylene. He reported no aseptic loosening. Callaghan et al 21 reported the results with 119 MBKA devices of the same design. There was no periprosthetic loosening or failure due to wear at a follow-up interval of 9 to 12 years. At intermediate follow- up of fully congruent multidirec- tional platforms used in 172 knees, Kaper et al 36 reported that only 2 (1.2%) required revision due to wear. However, these results are no better than those obtained with well- designed fixed-bearing TKA de- vices. 6,37 In the longer term, a clear difference in in vivo wear may be- come apparent as the fatigue limit of fixed-bearing TKA prostheses is reached. 38 An unanswered question remains about excessive volumetric wear in MBKA designs compared with fixed-bearing TKA designs. 39 Dual- articulation MBKA devices typically have a much larger articulating- surface contact area than fixed- bearing TKA devices, especially fully congruent designs. It is un- known whether, despite low linear wear rates, a greater or lesser vol- ume of particles is produced in vivo due to the larger contact area. Conflicting in vitro evidence exists concerning volumetric wear in MBKA. Finite-element modeling Christopher J. Vertullo, MBBS, FRACS(Ortho), et al Vol 9, No 6, November/December 2001 359 based on current theories of poly- mer failure suggests that low con- tact stress will result in minimal generation of abrasive wear parti- cles. 25 In an in vitro experimental model, Jones et al 39 suggested that multidirectional polyethylene plat- forms have wear rates nine times greater than unidirectional plat- forms. In a 10-million-cycle knee simulator study comparing a fully congruent multidirectional plat- form with a posteriorly stabilized fixed-bearing device, the multidi- rectional platform exhibited less linear wear than the fixed-bearing knee, but approximately 30% more volumetric wear. 15 Although negli- gible wear occurred at the femur- polyethylene surface, substantial abrasive and adhesive wear oc- curred at the tibia-polyethylene articulation. No delamination was observed at either surface. Further in vivo investigation is necessary in this area, especially regarding the role of multidirectional platforms and undersurface wear. It is unclear whether the optimal articulation geometry for MBKA is the fully congruent design with proximal contact stresses of less than 5 MPa over a surface area greater than 1,000 mm 2 or a gait- congruent MBKA design with prox- imal contact stresses of 5 to 8 MPa over a surface area of 500 mm 2 . Both designs can avoid the stress thresholds associated with delami- nation and pitting; however, the issue of minimizing linear wear ver- sus volumetric wear remains unre- solved. It has been suggested that MBKA is one solution to wear- particle generation from the under- surface of the polyethylene insert in TKA. This undersurface wear has been related to poor locking mechanisms in fixed-bearing TKA devices that allow micromotion against a rough tibial tray. 40 In the- ory, the easily manufactured, pol- ished bearing surface of an MBKA tibial tray can avoid this excessive backside wear. 9 However, no evi- dence exists that MBKA devices dis- play less backside wear than fixed- bearing TKA prostheses (some of which are also available with a pol- ished tibial tray). Two other design features of MBKA prosthesis may minimize undersurface wear. First, baseplate stiffness in MBKA devices typically exceeds that in conventional fixed- bearing devices. Mobile-bearing tibial plates constructed of cobalt- chrome alloys rather than titanium exhibit less deflection in load test- ing. This increased stiffness allows the baseplate to maintain even load distribution for the polyethylene. 15 Second, titanium tibial trays have poor wear characteristics that pre- clude their use in mobile-bearing knees in an untreated form. 36 Most current MBKA designs use cobalt- chrome alloys. Wear-particle generation from MBKA constraint mechanisms and from the tibial tray or polyethylene insert has been raised as a potential concern for mobile-bearing knees. 41 The results of recent experiments in which 1.5 million cycles of 400- to 800-N shear stress was applied to a multidirectional MBKA suggest that no plastic deformation of the polyethylene occurred due to the presence of tibial-baseplate stop mechanisms designed to limit poly- ethylene rotation and/or transla- tion. 15 However, malalignment and poor ligament balancing in MBKA could produce cyclic poly- ethylene impingement on con- straint stops in gait, producing excessive polyethylene particle generation. Rotating-platform devices toler- ate moderate amounts of tibial component malrotation compared with fixed-bearing TKA prosthe- ses. 36 An in vitro cadaver analysis showed lower femur–polyethylene surface peak stress in MBKA de- vices with 15 degrees of tibial malro- tation compared with fixed-bearing designs. 42 This advantage, however, does not obviate good surgical tech- nique. With severe malrotation (45 degrees), edge loading occurs, with the polyethylene bearing overhang- ing the tibial tray, markedly in- creasing undersurface stress. The argument that less wear debris is generated with an MBKA device than with a well-designed fixed-bearing TKA prosthesis has not been proved. 10 It must be re- membered that some fixed-bearing TKA knees have lower contact stresses in flexion than existing MBKA knees. 10,42 Reduction in Torque Forces An ideal TKA design maximizes articular conformity while minimiz- ing axial constraint. Constraint is the resistance to a particular degree of freedom, such as AP translation or axial rotation. 9 Conformity is a geometric measure of the closeness of fit of the knee articulation. In a flat-tibia TKA design, constraint is nominally zero, except for friction at the articulating surfaces; in a hinged TKA, the constraint is infinite. 9,30 Hinged knees and early highly con- strained TKA designs transferred ex- cessive axial torque or varus-valgus forces to the prosthesis-bone inter- face and tended to fail early. 35,43,44 Fixed-bearing TKA cannot be fully conforming without being exceed- ingly constrained to axial rotation, transferring large rotational stresses to the prosthesis-bone interface. Mobile-bearing devices can over- come this conformity–axial con- straint conflict by allowing uncon- strained axial rotation with fully conforming articulations, reducing the axial stress to the prosthesis- bone interface. 10,45 Unanswered questions remain about modern fixed-bearing de- signs that are more conforming Mobile Bearings in Primary Knee Arthroplasty Journal of the American Academy of Orthopaedic Surgeons 360 than earlier round-on-flat TKA prostheses. By increasing confor- mity to decrease wear, theoretically more axial torque is applied to the prosthesis-bone interface. 30,44 The threshold of axial torque stress at which prosthesis loosening occurs is unknown; however, minimizing this stress appears to be advanta- geous. Improvements in polyethyl- ene wear properties and quantifica- tion of the acceptable degree of constraint may allow fixed-bearing prostheses to outperform MBKA devices. Stability Ligamentous competence has a greater role in MBKA than in fixed- bearing TKA. Bicompartmental meniscal-bearing knees can retain either both cruciate ligaments or just the PCL. Higher failure rates for meniscal-bearing MBKA with an incompetent anterior cruciate liga- ment (ACL) or ACL sacrifice have been reported by a number of authors. 4,10,43,46,47 In one study, 4 a bearing fracture was related to ACL sacrifice that allowed posterior sub- luxation. However, other clinical studies have reported excellent results with an ACL-sacrificing mobile-bearing knee. 10 The production of anterior soft- tissue impingement by excessive anterior-platform translation of a multidirectional MBKA device has been reported. In one study of 16 cruciate-retaining multidirectional MBKA devices, 48 9 demonstrated unrestricted anterior translation of the platform with resultant im- pingement on the patellar tendon. The design lacked a stop mecha- nism for excessive translation. All 9 multidirectional platforms that al- lowed impingement had to be con- verted to a rotating platform. The authors theorized that PCL incom- petence in a cruciate-retaining de- sign results in excessive anterior translation. Cruciate-retaining mul- tidirectional MBKA devices are more reliant on PCL competence than similar fixed-bearing TKA devices, making successful recess- ing of a tight PCL imperative. 48 Stop mechanisms may decrease or prevent excessive AP translation in a multidirectional MBKA prosthesis in a patient with an incompetent PCL. 15 Matsuda et al, 20 in an in vitro cadaver study, compared stability in multidirectional MBKA, rotating MBKA, AP-translating MBKA, and fixed-bearing TKA prostheses and in normal knees. The multidirec- tional and AP-translating devices showed increased AP laxity in the absence of the ACL compared with normal knees and rotating MBKA and fixed-bearing TKA devices. The rotational stability of multidi- rectional and rotating MBKA de- vices was similar to that of normal knees; however, the fixed-bearing TKA showed decreased rotational deflection, which is evidence of rotational stress being transmitted to the prosthesis-bone interface. Hence, rotating and multidirectional MBKA devices have an advantage over fixed-bearing TKA prostheses in maintaining rotational stability while decreasing axial stress load. The authors also noted that malrota- tion of tibial components in fixed- bearing TKA results in a flexion contracture due to poor articular- surface mating. Dislocation and Breakage In addition to the accepted complica- tions associated with fixed-bearing TKA devices, the complexity of a mobile-bearing platform or menis- cus increases the chance of bearing subluxation or dislocation, bearing breakage, and soft-tissue impinge- ment. Bearing dislocations have been reported with a variety of MBKA designs; however, it must be noted that some large series have reported no bearing dislocations with either multidirectional 36 or rotating platforms. 21 Dislocation of rotating-only platforms was reported in 4 (9%) of 43 patients in one series. 43 However, the rate was only 0.15% in a much larger series (665 patients) involving the same MBKA design. 35 Reported rates of dislocation or subluxation with bicompartmen- tal meniscal-bearing devices have ranged between 2.2% 49 and 7.6%. 44 Dislocation rates of unicompartmen- tal meniscal bearings show a wide disparity between surgeons, hospi- tals, and countries. 10,50 From the lit- erature, it is evident a steep learning curve is associated with patient se- lection and surgical technique. 35 Smaller series have a much higher rate of complications. Many authors have emphasized good surgical technique to avoid bearing disloca- tion, especially balancing of flexion and extension gaps. 15,36,43 Many designs incorporate platform stop mechanisms that may reduce the risk of dislocation. Broken bearings appear to be more common with meniscal-bearing MBKA prostheses that utilize curved tracks without stops. Of the 16 pa- tients in one series, 4 4 (25%) had bro- ken lateral meniscal bearings, com- pared with 7 (1.5%) in a larger se- ries of 473 patients. 49 The authors theorized that the bearing breakage was due to posterior subluxation of the bearing and entrapment be- tween the femoral component and the posterior edge of the bearing track. Clinical Results The results of MBKA at intermedi- ate to long-term follow-up are equal to the best results reported for fixed- bearing TKA. Results at follow-up intervals in excess of 5 years are available for four prostheses: the Low Contact Stress Rotating Plat- Christopher J. Vertullo, MBBS, FRACS(Ortho), et al Vol 9, No 6, November/December 2001 361 form (DePuy), the Low Contact Stress Meniscal Bearing (DePuy), the Self Aligning Knee (Sulzer, Austin, Tex), and the Oxford Bi- compartmental Knee Replacement (Biomet). Callaghan et al 21 reported the 9- to 12-year follow-up data on 119 consecutive cruciate-sacrificing, cemented Low Contact Stress rotat- ing platform meniscal-bearing knees. No periprosthetic osteolysis or loosening was observed, and no revisions were required. The aver- age Hospital for Special Surgery scores improved from a preopera- tive value of 57 to 84 at follow-up. Average clinical and functional Knee Society scores improved from 30 and 44 preoperatively to 90 and 75 at follow-up. In another investigation, 49 the Kaplan-Meier survival estimate at 8 years for 473 consecutive cement- less cruciate-retaining Low Contact Stress Meniscal Bearing Knees was 94.6%. This investigation included a small percentage of bilaterally cruciate-preserving prostheses. Buechel and Pappas 51 have sup- ported the findings of Callaghan et al. 21 In their study of a rotating plat- form device, the 10-year survival rate for cemented prostheses was 97.5%, and the 6-year survival rate for cementless designs was 98.1%. 51 Sorrells 35 also reported favorable results with the Low Contact Stress Rotating Platform prosthesis. In a series of consecutive noncemented, PCL-sacrificing Low Contact Stress Rotating Platform implants, the rate of a good or excellent outcome at 1 to 11 years was greater than 98%, and the survivorship at 11 years was 95%. Results with the Oxford Bicom- partmental meniscal-bearing knee have been similar, provided the ACL remains intact. 46,52 At the 6- year follow-up, the success rate was 93% for patients with intact ACL function, compared with 73% when the ACL was compromised. Kaper et al 36 reported the results with 172 PCL-retaining multidirec- tional Self Aligning Knee prostheses. The mean follow-up interval was 5.6 years (range, 5 to 8 years). Of the 141 patients, 132 (94%) described their re- sults as good or very good. Two re- vision procedures were necessitated by polyethylene wear, but none of the remaining knees showed evi- dence of wear. Kaplan-Meier sur- vival curves showed the probability of survival to be 91.7% with revision surgery for any reason as an end point and 98.8% for revision surgery because of polyethylene wear as an end point. Short-term follow-up for other mobile-bearing TKA devices appears promising, 15 but longer follow-up for some of these designs remains unpublished. Furthermore, these studies involved older patients (aged 64 to 70 years); no results are avail- able for mobile-bearing TKA per- formed in cohorts limited to younger, more active patients. The results of MBKA have been similar to those obtained with well- designed fixed-bearing TKA devices. Investigators have reported 95% good-to-excellent results and implant survival rates consistently greater than 94% with conventional fixed- bearing TKA devices at a follow-up interval of 10 to 15 years. 5-7,37 Re- ports of fixed-bearing TKA in pa- tients less than 55 years are en- couraging 53-55 ; however, the average follow-up interval was less than 10 years. Indications and Limitations The literature on the indications for MBKA is scant. Theoretically, MBKA may offer improved kine- matics and reduced wear to the rela- tively young patient; however, no data exist to support this hypothesis. It is clear that not all patients are candidates for MBKA. Obviously, severe malalignment and ligamen- tous incompetence necessitate the use of constrained fixed-bearing knees, especially in the elderly. 56 However, the degree of deformity that can be treated with MBKA is unclear. Marked fixed flexion defor- mities make accurate soft-tissue bal- ancing of the knee while retaining the PCL technically difficult. In this situation, multidirectional platforms that rely on the PCL to avoid exces- sive anterior translation would be relatively contraindicated. With MBKA-only systems, the surgeon must decide preoperatively which knee system to use or accept the added cost of having both a fixed-bearing and a mobile-bearing system available intraoperatively. Total knee arthroplasty systems that allow the surgeon intraoperative choice in the type of bearing to use are more flexible; however, none of these systems has FDA approval, nor are there intermediate clinical results available. No clear indications yet exist as to the appropriate clinical applica- tions for the large variety of MBKA permutations currently available. Rotating platforms offer greater AP stability and less reliance on PCL function compared with PCL- retaining multidirectional plat- forms, while being rotationally unconstrained and self-aligning. Multidirectional platforms with stop mechanisms appear to over- come this concern. Summary The MBKA designs have theoreti- cal advantages over fixed-bearing designs due to their ability to de- crease axial constraint while pro- viding less linear wear by virtue of increased conformity and self- correction of tibial component mal- rotation. However, MBKA devices have disadvantages related to bear- ing dislocation and breakage, soft- Mobile Bearings in Primary Knee Arthroplasty Journal of the American Academy of Orthopaedic Surgeons 362 tissue impingement, technique learn- ing curve, and concerns about volu- metric wear. Current MBKA designs have kinematic abnormalities similar to those that fixed-bearing knees exhibit. The long-term clinical results with both fixed-bearing and par- tially congruent rotating-platform MBKA devices are excellent, al- though the theoretical benefits of MBKA have not yet improved the clinical results compared with fixed- bearing TKA. Hypothetically, longer-term follow-up may reveal a statistically significant difference as the fatigue failure threshold of fixed- bearing knees is exceeded, 37 but there are as yet no clinical data. Surgeons willing to accept the disad- vantages of MBKA can continue to do so, knowing that the current clin- ical results of MBKA match those obtained with the best fixed-bearing TKA devices, with the chance of improved survival past 20 years. Mobile-bearing designs offer a confusing array of options, includ- ing degree of conformity, constraint mechanisms, directional mobility of the bearing, and PCL management. Results at follow-up intervals longer than 10 years are available only for partially congruent, PCL-sacrificing rotating platforms and PCL-retaining meniscal-bearing designs. The in- termediate results for multidirection- al MBKA are promising; however, longer-term results are needed to overcome concerns about AP instabil- ity and volumetric wear. Evidence- based, stepwise introduction of new orthopaedic devices allows safe and controlled implementation of new technologies while exposing as few patients as possible to the risk of fail- ure. 57 When used as part of a TKA system, MBKA appears to allow re- duction in inventory and offer greater intraoperative flexibility. A mobile bearing will not com- pensate for poor surgical technique or basic TKA design flaws. Al- though platforms are rotationally self-aligning, this does not allow decreased attention to surgical tech- nique. Avoidance of bearing dislo- cation and breakage is dependent on balanced flexion and extension gaps. In summary, MBKA is theoretically attractive, but there are as yet no data indicating that it is superior to fixed-bearing TKA. Christopher J. Vertullo, MBBS, FRACS(Ortho), et al Vol 9, No 6, November/December 2001 363 References 1. O’Connor JJ, Goodfellow JW: Theory and practice of meniscal knee replace- ment: Designing against wear. Proc Inst Mech Eng [H] 1996;210:217-222. 2. Menchetti PPM, Walker PS: Mechani- cal evaluation of mobile bearing knees. Am J Knee Surg 1997;10:73-82. 3. Morra EA, Postak PD, Greenwald AS: The influence of mobile bearing knee geometry on the wear of ultra-high molecular weight polyethylene tibial inserts: A finite element study [exhibit]. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons, Anaheim, Calif, February 4-8, 1999. 4. Weaver JK, Derkash RS, Greenwald AS: Difficulties with bearing disloca- tion and breakage using a movable bearing total knee replacement system. Clin Orthop 1993;290:244-252. 5. Colizza WA, Insall JN, Scuderi GR: The posterior stabilized total knee prosthesis: Assessment of polyethyl- ene damage and osteolysis after a ten- year-minimum follow-up. J Bone Joint Surg Am 1995;77:1713-1720. 6. Font-Rodriguez DE, Scuderi GR, Insall JN: Survivorship of cemented total knee arthroplasty. Clin Orthop 1997; 345:79-86. 7. Ranawat CS, Flynn WF Jr, Saddler S, Hansraj KK, Maynard MJ: Long-term results of the total condylar knee arthroplasty: A 15-year survivorship study. Clin Orthop 1993;286:94-102. 8. Insall JN: Adventures in mobile- bearing knee design: A mid-life crisis. Orthopedics 1998;21:1021-1023. 9. Walker PS, Sathasivam S: Design forms of total knee replacement. Proc Inst Mech Eng [H] 2000;214:101-119. 10. Callaghan JJ, Insall JN, Greenwald AS, et al: Mobile-bearing knee replace- ment: Concepts and results. J Bone Joint Surg Am 2000;82:1020-1041. 11. Hollister AM, Jatana S, Singh AK, Sullivan WW, Lupichuk AG: The axes of rotation of the knee. Clin Orthop 1993;290:259-268. 12. Churchill DL, Incavo SJ, Johnson CC, Beynnon BD: The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop 1998;356: 111-118. 13. Walker PS, Sathasivam S: Controlling the motion of total knee replacements using intercondylar guide surfaces. J Orthop Res 2000;18:48-55. 14. D’Lima DD, Trice M, Urquhart AG, Colwell CW Jr: Comparison between the kinematics of fixed and rotating bearing knee prostheses. Clin Orthop 2000;380:151-157. 15. Insall JN, Aglietti P, Baldini A, Easley ME: Meniscal-bearing knee replace- ment, in Insall JN, Scott WN (eds): Surgery of the Knee, 3rd ed. New York: Churchill Livingstone, 2001, pp 1717- 1738. 16. Stiehl JB, Dennis DA, Komistek RD, Keblish PA: In vivo kinematic analy- sis of a mobile bearing total knee pros- thesis. Clin Orthop 1997;345:60-66. 17. Callahan CM, Drake BG, Heck DA, Dittus RS: Patient outcomes following tricompartmental total knee replace- ment: A meta-analysis. JAMA 1994; 271:1349-1357. 18. Dennis DA, Komistek RD, Stiehl JB, Walker SA, Dennis KN: Range of motion after total knee arthroplasty: The effect of implant design and weight-bearing conditions. J Arthro- plasty 1998;13:748-752. 19. Stiehl JB, Voorhorst PE, Keblish P, Sorrells RB: Comparison of range of motion after posterior cruciate liga- ment retention or sacrifice with a mobile bearing total knee arthroplasty. Am J Knee Surg 1997;10:216-220. 20. Matsuda S, Whiteside LA, White SE, McCarthy DS: Knee stability in menis- cal bearing total knee arthroplasty. J Arthroplasty 1999;14:82-90. 21. Callaghan JJ, Squire MW, Goetz DD, Sullivan PM, Johnston RC: Cemented rotating-platform total knee replace- ment: A nine to twelve- year follow-up study. J Bone Joint Surg Am 2000;82: 705-711. 22. Bartel DL, Bicknell VL, Wright TM: The effect of conformity, thickness, and material on stresses in ultra-high molecular weight components for total joint replacement. J Bone Joint Surg Am 1986;68:1041-1051. 23. McGloughlin TM, Kavanagh AG: Wear of ultra-high molecular weight polyethylene (UHMWPE) in total knee prostheses: A review of key in- fluences. Proc Inst Mech Eng [H] 2000; 214:349-359. 24. Lewis G: Design issues in clinical studies of the in vivo volumetric wear rate of polyethylene bearing compo- nents. J Bone Joint Surg Am 2000;82: 281-287. 25. Morra EA, Postak PD, Greenwald AS: The Influence of Mobile Bearing Knee Geometry on the Wear of UHMWPE Tibial Inserts: II. A Finite Element Study. Cleveland, Ohio: Orthopaedic Research Laboratories, 1999. 26. Rostoker W, Galante JO: Contact pres- sure dependence of wear rates of ultra high molecular weight polyethylene. J Biomed Mater Res 1979;13:957-964. 27. Morra EA, Postak PD, Greenwald AS: The effects of articular geometry on delamination and pitting of UHMWPE tibial inserts: I. A finite element study. Orthop Trans 1996-1997;20:66. 28. Blunn GW, Joshi AB, Minns RJ, et al: Wear in retrieved condylar knee arthro- plasties: A comparison of wear in dif- ferent designs of 280 retrieved condylar knee prostheses. J Arthroplasty 1997;12: 281-290. 29. Kuster MS, Horz S, Spalinger E, Stachowiak GW, Gächter A: The effects of conformity and load in total knee re- placement. Clin Orthop 2000;375:302-312. 30. Werner F, Foster D, Murray DG: The influence of design on the transmis- sion of torque across knee prostheses. J Bone Joint Surg Am 1978;60:342-348. 31. Buechel FF, Pappas MJ: The New Jersey Low-Contact-Stress Knee Re- placement System: Biomechanical rationale and review of the first 123 cemented cases. Arch Orthop Trauma Surg 1986;105:197-204. 32. Morra EA, Postak PD, Greenwald AS: The effects of articular geometry on delamination and pitting of UHMWPE tibial inserts: II. A finite element study. Orthop Trans 1997-1998;21:217. 33. Argenson JN, O’Connor JJ: Polyethyl- ene wear in meniscal knee replace- ment: A one to nine-year retrieval analysis of the Oxford knee. J Bone Joint Surg Br 1992;74:228-232. 34. Psychoyios V, Crawford RW, O’Connor JJ, Murray DW: Wear of congruent me- niscal bearings in unicompartmental knee arthroplasty: A retrieval study of 16 specimens. J Bone Joint Surg Br 1998; 80:976-982. 35. Sorrells RB: The rotating platform mo- bile bearing TKA. Orthopedics 1996;19: 793-796. 36. Kaper BP, Smith PN, Bourne RB, Rora- beck CH, Robertson D: Medium term results of a mobile bearing total knee replacement. Clin Orthop 1999;367: 201-209. 37. Gill GS, Joshi AB, Mills DM: Total con- dylar knee arthroplasty: 16- to 21-year results. Clin Orthop 1999;367:210-215. 38. Li EC, Ritter MA, Montgomery T, Fur- man BD, Li S, Wright TM: Catastrophic failure of a conforming type of total knee replacement: A case report. Clin Orthop 1996;333:234-238. 39. Jones VC, Barton DC, Fitzpatrick DP, Auger DD, Stone MH, Fisher J: An experimental model of tibial counter- face polyethylene wear in mobile bear- ing knees: The influence of design and kinematics. Biomed Mater Eng 1999;9: 189-196. 40. Wasielewski RC, Parks N, Williams I, Surprenant H, Collier JP, Engh G: Tibial insert undersurface as a con- tributing source of polyethylene wear debris. Clin Orthop 1997;345:53-59. 41. Scott RD: Mobile or Fixed Plateau: What’s the Answer? Cleveland, Ohio: Orthopaedic Research Laboratories, 2000. 42. Matsuda S, White SE, Williams VG II, McCarthy DS, Whiteside LA: Contact stress analysis in meniscal bearing total knee arthroplasty. J Arthroplasty 1998;13:699-706. 43. Bert JM: Dislocation/subluxation of meniscal bearing elements after New Jersey low-contact stress total knee ar- throplasty. Clin Orthop 1990;254:211-215. 44. Walker PS: Requirements for success- ful total knee replacements: Design considerations. Orthop Clin North Am 1989;20:15-29. 45. Heim CS, Postak PD, Greenwald AS: Mobility Characteristics of Mobile Bearing Total Knee Designs. Cleveland, Ohio: Orthopaedic Research Laboratories, 1999. 46. Goodfellow J, O’Connor J: The anterior cruciate ligament in knee arthroplasty: A risk-factor with unconstrained me- niscal prostheses. Clin Orthop 1992;276: 245-252. 47. White SH, O’Connor JJ, Goodfellow JW: Sagittal plane laxity following knee arthroplasty. J Bone Joint Surg Br 1991;73:268-270. 48. Walsh WR: Instability of an anteropos- terior glide mobile bearing knee. Pre- sented at the Australian Orthopaedic Association Annual Scientific Meeting, Brisbane, 1999. 49. Jordan LR, Olivo JL, Voorhorst PE: Survivorship analysis of cementless meniscal bearing total knee arthroplas- ty. Clin Orthop 1997;338:119-123. 50. Lewold S, Goodman S, Knutson K, Robertsson O, Lidgren L: Oxford meniscal bearing knee versus the Marmor knee in unicompartmental arthroplasty for arthrosis: A Swedish multicenter survival study. J Arthro- plasty 1995;10:722-731. 51. Buechel FF, Pappas MJ: Long-term sur- vivorship analysis of cruciate-sparing versus cruciate-sacrificing knee pros- theses using meniscal bearings. Clin Orthop 1990;260:162-169. 52. Goodfellow JW, O’Connor J: Clinical results of the Oxford knee: Surface arthroplasty of the tibiofemoral joint with a meniscal bearing prosthesis. Clin Orthop 1986;205:21-42. 53. Diduch DR, Insall JN, Scott WN, Scuderi GR, Font-Rodriguez D: Total knee replacement in young, active patients: Long-term follow-up and functional outcome. J Bone Joint Surg Am 1997;79:575-582. 54. Ranawat CS, Padgett DE, Ohashi Y: Total knee arthroplasty for patients younger than 55 years. Clin Orthop 1989;248:27-33. 55. Lonner JH, Hershman S, Mont M, Lotke PA: Total knee arthroplasty in patients 40 years of age and younger with osteoarthritis. Clin Orthop 2000; 380:85-90. 56. Easley ME, Insall JN, Scuderi GR, Bullek DD: Primary constrained con- dylar knee arthroplasty for the ar- thritic valgus knee. Clin Orthop 2000; 380:58-64. 57. Malchau H: Introducing new technol- ogy: A stepwise algorithm [editorial]. Spine 2000;25:285. Mobile Bearings in Primary Knee Arthroplasty Journal of the American Academy of Orthopaedic Surgeons 364 . minimizing this stress appears to be advanta- geous. Improvements in polyethyl- ene wear properties and quantifica- tion of the acceptable degree of constraint may allow fixed-bearing prostheses to outperform