Bertrand W. Parcells, MD, and Alfred J. Tria Jr., MD
Authors’ Disclosure Statement: Dr. Tria reports that he receives royalties from Smith & Nephew, and is a consultant for Smith & Nephew, Medtronic, and Pacira. Dr. Parcells reports no actual or potential conflict of interest in relation to this article.
The early knee replacements were hinge designs that ignored the ligaments of the knee and resurfaced the joint, allowing freedom of motion in a single plane. Advances in implant fixation paved the way for modern designs, including the posterior-stabilized (PS) total knee arthroplasty (TKA) that sacrifices both cruciate ligaments while substituting for the posterior cruciate ligament (PCL), and the cruciate-retaining (CR) TKA designs that sacrifice the anterior cruciate ligament but retain the PCL. The early bicruciate retaining (BCR) TKA designs suffered from loosening and early failures. Townley and Cartier designed BCR knees that had better clinical results but the surgical techniques were challenging.
Kinematic studies suggest that normal motion relies on preservation of both cruciate ligaments. Unicompartmental knee arthroplasty retains all knee ligaments and closely matches normal motion, while PS and CR TKA deviate further from normal. The 15% to 20% dissatisfaction rate with current TKA has renewed interest in the BCR design. Replication of normal knee kinematics and proprioception may address some of the dissatisfaction.
Hinge knee arthroplasty was introduced in the 1950s.1 All 4 major ligaments were replaced by the hinge, which provided stabilization while allowing sagittal plane motion. Its goal was stability, not replication of normal kinematics. The addition of methyl methacrylate cement improved fixation and allowed surface design modifications that addressed normal articular motion. Implants such as the Gunston Polycentric,2 the Duocondylar,3 and the Geometric4 resurfaced the medial and lateral compartments of the knee while preserving the cruciate ligaments. The implants were subject to greater translational forces without the hinge and loosening became a major problem despite the advances in cementing. It became evident in the 1970s that preservation of the cruciates complicated the procedure. Cruciate resection simplified the operation and allowed improved fixation. The ICLH prosthesis resected the cruciates and used the articular surface design to give stability to the knee.5,6 The total condylar prosthesis had a “tibial” imminence that mimicked the shape of the tibial surface but also sacrificed both of the cruciate ligaments (Figure 1).
Designers recognized that the cruciate ligaments affected knee kinematics; however, they elected to sacrifice the anterior cruciate ligament (ACL) for surgical simplicity and implant longevity.6 In the early 1980s, both the cruciate-retaining (CR) total knee arthroplasty (TKA) (Figure 2) and posterior-stabilized (PS) TKA (Figure 3) designs addressed the posterior cruciate ligament (PCL) function. The PCL was preserved in the “cruciate-retaining” TKA, substituted in the “posterior-stabilized” TKA using a cam-post mechanism. The CR TKA designers believed that PCL preservation produced a more balanced knee with a more anatomical result, a more normal joint line, and better function, especially on stair climbing. The PS TKA designers admitted the value of posterior stabilization but argued that it was too difficult to consistently save the PCL in all cases, and that the PS knee was easier for surgeons to implant with more reliable roll back.7
The Geometric knee was developed in the 1970s to retain both cruciate ligaments.4 Unfortunately, it created a kinematic conflict by using a constrained articular surface design that prevented the motion required by the cruciate ligaments. This conflict resulted in tibial loosening and early failures. The compromised results decreased interest in the bicruciate-retaining (BCR) TKA designs, allowing the CR TKA and PS TKA designs to flourish for the next 20 years with little or no attempts to retain the ACL.
In the 1980s the BCR TKA design was pursued by Townley8 and Cartier.9 Townley8 believed that cruciate resection was a concession to “improper joint synchronization”8 and Cartier9 thought that cruciate preservation permitted more normal proprioception.9 Unlike prior BCR TKA designs, the mid-term clinical results were equal to or better than the standard CR TKA or PS TKA of the time, and 9- to 11-year follow-up demonstrated comparable outcomes.8 While these results highlighted the possibility of a BCR TKA, the surgical technique and failures of the Geometric knee discouraged surgeons from pursuing the BCR TKA.
Interest in cruciate-preserving knee arthroplasty returned with partial knee replacements, with patients reporting more normal proprioception and motion.10 The techniques became more popular with the introduction of the minimally invasive surgeries in the early 2000s and cruciate ligament preservation became a more interesting concept.11,12 Some surgeons preserved the cruciates by using separate implants for the medial, lateral, and patellofemoral surfaces.10 These results were acceptable for the time but required considerable surgical talent and did not report 20-year results similar to the CR and PS knees.
Most prosthetic designs attempt to copy the normal knee anatomy. Using fluoroscopic studies and computer analysis, designers began to investigate the motion (or kinematics) of the normal knee and realized that despite the fact the TKA looked like the human knee, the designs were not kinematically correct.13
Although TKA successfully treats pain secondary to degenerative joint disease, many patients are unable to return to their prior level of function, with up to 20% reporting dissatisfaction with their level of activity.14 The observed differences in kinematics between a normal knee and a TKA may explain part of this discrepancy.
Normal Knee Motion
The tibiofemoral articulation in a normal knee follows a reproducible pattern of motion as the knee moves from extension to flexion. The lateral femoral condyle (LFC) translates posteriorly with a combination of rolling and sliding motion, while the medial femoral condyle (MFC) has minimal posterior translation and thus acts as a pivot for knee motion. The MFC is larger, less curved, and has a biphasic shape with 2 distinct radiuses of curvature that correspond to an “extension” and “flexion” facet. The transition between the MFC facets occurs at approximately 30° of flexion, whereby the contact point transfers posteriorly with little condylar translation.15-17 In contrast, the LFC is smaller, has a single radius of curvature, and gradually translates posteriorly throughout flexion. Static magnetic resonance imaging of the knee from 0° to 120° shows an average of 19 mm posterior translation for the LFC and 2 mm for the MFC.15-20
