Fixed or mobile-bearing total knee arthroplasty
© Huang et al; licensee BioMed Central Ltd. 2007
Received: 28 August 2006
Accepted: 05 January 2007
Published: 05 January 2007
Skip to main content
© Huang et al; licensee BioMed Central Ltd. 2007
Received: 28 August 2006
Accepted: 05 January 2007
Published: 05 January 2007
Fixed and mobile-bearing in total knee arthroplasty are still discussed controversially. In this article, biomechanical and clinical aspects in both fixed and mobile-bearing designs were reviewed. In biomechanical aspect, the mobile-bearing design has proved to provide less tibiofemoral contact stresses under tibiofemoral malalignment conditions. It also provides less wear rate in in-vitro simulator test. Patients with posterior stabilized mobile-bearing knees had more axial tibiofemoral rotation than patients with posterior stabilized fixed-bearing knees during gait as well as in a deep knee-bend activity. However, in clinical aspect, the mid-term or long-term survivorship of mobile-bearing knees has no superiority over that of fixed-bearing knees. The theoretical advantages for mobile-bearing design to provide a long-term durability have not been demonstrated by any outcome studies. Finally, the fixed-bearing design with all-polyethylene tibial component is suggested for relatively inactive, elder people. The mobile-bearing design is suggested for younger or higher-demand patients due to the potential for reduced polyethylene wear and more normal kinematics response after joint replacement. For younger surgeon, the fixed-bearing design is suggested due to less demand for surgical technique. For experienced surgeon, one familiar surgical protocol and instrumentation is suggested rather than implant design, either fixed-bearing or mobile-bearing.
Total knee arthroplasty (TKA) has become a standard operative procedure to relieve pain and restore function in patients with osteoarthritis or rheumatoid arthritis. Current total knee prosthesis (TKP) devices can be subdivided into two groups based on different fundamental design principals: fixed-bearing knees, where the polyethylene tibial insert locked with tibial tray, and mobile-bearing designs which facilitate movement of the insert relative to the tray . In each group, the knee system can be further subdivided into three groups, posterior cruciate ligament retained, sacrificed or substituted. In mobile-bearing group, some designs allow both anterior-posterior translation and internal-external rotation at the tray-insert interface while others rotation only is facilitated at the tray-insert counterface .
Implant loosening and polyethylene wear in fixed-bearing knee prostheses were recognized as major causes of late failure. Fixed-bearing prosthesis with a high conformity bearing surface provides low contact stress, but produces high torque at the bone-implant interface predisposing to component loosening. Conversely, prosthesis with a low conformity bearing surface produces less constraint force that decreasing component loosening, but generates high contact stress leading to early failure of the polyethylene [2, 3]. Furthermore, the kinematic conflict between low-stress articulations and free rotation cannot be solved by any fixed-bearing knee design .
Mobile-bearing knee prosthesis was introduced with the aim to reduce polyethylene wear and component loosening . The mobile-bearing design provides both congruity and mobility in the tibiofemoral bearing surface. This allows low contact stress and low constraint force to improve wear resistance and, theoretically, to minimize loosening . In addition, the mobile-bearing knee also solves the kinematic conflict of fixed-bearing knee because a high conforming articular surface can now coexist with free rotation .
Although mobile-bearing design has hypothetical advantages over fixed-bearing knee, both designs show excellent survival rates of up to 95% in 10-year follow-up [6–11] and comparative studies could not demonstrate the superiority of one or the other design [10, 12–14]. It appears we are at a crossroad. In this article, two aspects including biomechanical and clinical analysis in both fixed and mobile-bearing designs were reviewed.
One major goal of mobile-bearing knee is to reduce the overall wear damage by increasing the contact area, while minimizing the constraint and encouraging nature knee motion, by allowing the polyethylene bearings to move freely on polished plates on the upper tibia . McEwen et al.  used a physiological knee simulator to compare the wear rate of fixed-bearing and rotating platform mobile-bearing TKP. The PFC Sigma rotating platform mobile-bearing knee (Depuy, Warsaw, IN) exhibited a mean wear rate of 5.2 ± 3.8 mm3/million cycles when subjected to standard kinematics with force-controlled anterior-posterior translation. This showed a two-fold reduction in volumetric wear rate in comparison to the fixed-bearing PFC Sigma knees subjected to intermediate kinematics (9.8 ± 3.7 mm3/million cycles). The authors  attributed the lower wear rate in mobile-bearing design to that rotating platform mobile-bearing design redistributes the motions between the femoral-insert and tray-insert articular surfaces. Most of the rotation occurs at the tray-insert articular surface, which is simply a unidirectional rotation motion and is known to produce low wear [20, 21]. In contrast, rotation of the PFC Sigma fixed-bearing knee occurs entirely at the femoral-insert articulation. The resulting multidirectional wear path at the interface increases the amount of cross shear on the polyethylene articulating surface, therefore, produces a greater polyethylene wear rate . However, lower wear rate cannot represent lower risk of osteolysis in vivo because the tissue response is depended on the size distributions of wear particles. When the particle size from different types of knee prosthesis were measured, the less conforming the design, the higher the surface damage, but the larger the particle size . Mobile-bearing knee may have a higher rate of production of smaller particles than fixed-bearing knee due to its larger contact area. This hypothesis was disproved by an in vitro pin-on-disk wear test . In that study, the authors showed that the wear rates of knee prosthesis with large contact areas, such as mobile-bearing designs, can be much less than that of fixed-bearing designs. However, there is no disadvantage regarding particle type or size associated with the larger contact areas.
In vitro cadaveric kinematics study under controlled laboratory condition was conducted by Most et al . In their study, eleven human knee specimens retrieved post-mortem were tested using a robotic system. The tibiofemoral translation and rotation of the intact and two reconstructed knees, fixed-bearing posterior stabilizing knee (LPS-Flex, Zimmer, IN) and mobile-bearing posterior stabilized knee (LPS-Mobile, Zimmer, IN), were compared. One force-moment control algorithm  to determine the passive path from full extension to 120° and three variations of muscle loads were simulated in that study: (1) isolated quadriceps force of 400 N; (2) combined quadriceps (400 N) and hamstrings (200 N) load; and (3) isolated hamstring force of 200 N. The kinematics of the intact and reconstructed knees under these simulated muscles loads was measured at selected flexion angles (0°, 30°, 60°, 90° and 120°). Their results indicated that for all knees posterior femoral translation occurs along the passive path and under muscle loading conditions. Furthermore, increasing flexion angle corresponded with increased internal tibial rotation. Femoral translation and tibial rotation for fixed- and mobile-bearing posterior stabilized knees were similar despite component design variations. However, both reconstructed knees only partially restored intact knee translation and rotation.
In vivo kinematics studies by using fluoroscopy have been conducted on subjects with normal knee joints and on patients who had implantation of a fixed-bearing or mobile-bearing TKP [25–30]. The most important parameters, the anterior-posterior translation and axial tibiofemoral rotation, have been determined in some studies of Low Contact Stress (LCS) posterior-cruciate-sacrificing rotating platform and posterior stabilized rotating platform mobile-bearing knees (Depuy, Warsaw, IN) [25, 26]. In normal knee's kinematics as the subject performed a weight-bearing deep knee-bend, there is posterior translation (14.2 mm) of lateral femoral condyle with progressive knee flexion . Contact pathways in patients who had a posterior-cruciate-retaining fixed-bearing knee showed a small amount of posterior femoral rollback (4.8 mm) in the lateral component occurred during the first 60 degrees of flexion, following by anterior femoral translation as the knee flexed from 60 to 90 degrees . Patients with a rotating platform knees had minimal anteroposterior tibiofemoral translation during a deep knee-bend, with tibiofemoral contact point remaining near the middle of the articular surface of tibial component .
Patients who had a posterior stabilized fixed-bearing TKA routinely demonstrated posterior femoral rollback in the lateral component during knee flexion, although it was less in magnitude than that in the normal knees . However, patients who had a posterior-cruciate-retaining fixed bearing knee experienced a paradoxical anterior femoral translation during a deep knee-bend . On the other hand, patients managed with a posterior-cruciate-sacrificing rotating platform knee replacement had posterior femoral rollback of the lateral femoral condyle from full extension to 90 degrees of flexion, but they actually experienced anterior translation from 60 to 90 degrees of flexion . Patient who had a posterior stabilized rotating platform knee replacement showed more substantial posterior femoral rollback of the lateral condyle during the deep knee-bend .
Axial tibiofemoral rotation after TKA is another important parameter for knee's kinematics. A multicenter analysis to determine in vivo axial tibiofemoral rotational magnitude and patterns in 1,027 knees was reported . In that study, normal knees showed 16.5° and 5.7° of internal tibial rotation during a deep knee-bend and gait, respectively. Patients with a posterior stabilized mobile-bearing knee prosthesis had, on average, more rotation (2.2°) than patients with a posterior stabilized fixed-bearing knee (1.4°) during the stance-phase of gait. However, patients with posterior-cruciate-retaining fixed-bearing (2.1°) knee had less rotation than patients with a posterior-cruciate-retaining mobile-bearing knee (0.1°). In a deep knee-bend activity, patients with posterior stabilized mobile-bearing design had a mean rotation of 3.9° while patients with posterior stabilized fixed-bearing one had 3.1° of rotation. Patients with posterior-cruciate-retaining mobile-bearing knee had a mean rotation of 3.9° while patients with posterior-cruciate-retaining fixed-bearing one had 3.7° of rotation. In addition, Ranawat et al  also investigated the tibiofemoral rotation in mobile-bearing PCL-sacrificed knee and fixed-bearing posterior stabilized knee during deep knee-bend activity. Although their results indicated that patients with a mobile-bearing prosthesis (7.3°) experienced a statistically greater amount of axial rotation than patients with a fixed-bearing prosthesis (4.1°), it is difficult to say the mobile-bearing knee has a greater axial rotation than fixed-bearing knee in vivo due to different design concept for implants, ie, PCL sacrificed vs posterior stabilized. Based on above-mentioned studies, patients with any design of knee prostheses (fixed-bearing or mobile-bearing) had smaller femoral rollback as well axial rotation than normal knees during gait and a deep knee bend. In addition, high variability in femoral rollback and axial rotation patterns and magnitudes were seen among different TKA designs.
Some conventional fixed-bearing TKAs have been proved to be clinical successful. Survivorship of the Genesis (Smith and Nephew, Memphis, TN) TKA was 96% at 10 years follow-up . Ritter et al  reported a survivorship of 98.8% at 15 years with the Anatomic Graduated Components (Biomet, Warsaw, IN) TKA. The survival rate of the Total Condyle knee prostheses (Howmedica, Rutherford, JN) was 95% at 15 years, 98% at 20 years and 91% at 23 years in different studies [49–51]. In our experience , using revision for mechanical failure as an end point, the 20-year overall survival rate with the Total Condyle knee prostheses was 91.9%. We also found the survival rate for the all-polyethylene tibial component was 96.4% and for the metal-backed tibial component was 88.4%. The use of more-cost effective and durable all-polyethylene tibial component for a primary cemented TKA, particularly in Asians with a relatively low weight and inactive, especially in elderly people is suggested.
However, in the current situation, surgeons face more and more young, active patients who need TKA. Patient's expectations for a more functional and longer-lasting result following TKAs continue to drive advances in both implant design as well as the surgical technique. The mobile-bearing knee prosthesis was designed to reduce contact stress and constraint force, which potentially, provided long-term durability. Buechel et al  reported a 20-year survival rate of the LCS cemented rotating platform prosthesis of 97.7% and a 16-year survival rate for the LCS cementless meniscal-bearing of 83%. Jordan et al  reported the survivorship of the meniscal-bearing prosthesis was 94.8% at 8 years. In our series, 495 primary LCS TKAs was reviewed . Among them, 228 knees were with meniscal-bearing prostheses and the remaining 267 knees were with rotating platform. The mean follow-up was 12 years (range 10 to 15 years). The overall survivorship was 88.1% at 15 years using Kaplan-Meier analysis. The survival rate was 83% for the meniscal-bearing prostheses and 92.1% for the rotating-platform prostheses. The mobile-bearing knee prosthesis has no superiority over that of fixed-bearing knees, especially for the mensical-bearing design.
Although long-term survivorship for fixed-bearing and mobile-bearing knees have been reported, there are few studies to compare the performance of fixed-bearing and mobile-bearing in patients with bilateral TKAs. Bhan et al  reported their series in 32 patients who had bilateral knee arthritis with similar deformity and preoperative range of motion on both sides. Patients agreed to have one knee replaced with a mobile-bearing knee and the other with a fixed-bearing one. In a minimum follow-up of 4.5 years, the results showed that no benefit of mobile-bearing over fixed-bearing designs could be demonstrated with respect to Knee Society scores, range of motion, subjective preference or patellofemoral complication rates. The risk of bearing subluxation and dislocation in knees with the mobile-bearing prosthesis is a cause for concern and may necessitate early revision . In another study which compared the mid-term follow-up of mobile-bearing and fixed-bearing in bilateral TKAs , the result also showed no difference between mobile-bearing and fixed-bearing prostheses. So far, the theoretical advantages for mobile-bearing design to provide a long-term durability have not been demonstrated by any outcome studies.
The exact indications for using the mobile-bearing knee are still unclear. Even though the mobile-bearing design has theoretically favourable features compared with fixed-bearing system, these have not been proven biomechanically and to extend the implant longevity in clinical aspect. Based on our experience in the past thirty-year, the fixed-bearing design with all-polyethylene tibial component is suggested for relatively inactive, elder people. For younger or higher-demand patients, the mobile-bearing design is suggested due to the potential for reduced polyethylene wear and more normal kinematics response after joint replacement. For the younger surgeon, using the fixed-bearing design is suggested due to less demand in surgical technique. For the experienced surgeon, one familiar surgical protocol and instrumentation is suggested rather than implant design, either in fixed-bearing or mobile-bearing.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.