Our findings can be summarized as follows: (a) the patellar ridge was an oblique line from the proximal to the distal medial patella, which located in the medial patellar facet, and (b) the center of the patellar ridge could be located in three areas with regard to the centre of the patellar cut: superiorly and medially in 88.75%, inferiorly and medially in 8.75%, laterally and superiorly in 2.5%.The most important finding of this study was that the newly developed CT-based 3D computer model could be useful for measuring and virtually resecting the patella. The anthropometric measurements of patellar ridge could provide basic data for guiding surgical management of the patella in TKA and are useful in designing patellar implants.
The function of the patella is to increase the lever arm of the extensor mechanism and make knee flexion more effective. It connects with the trochlear groove through its vertical ridge, dividing the patella into medial and lateral facets[16]. Morphological analyses of the patella such as width and thickness of the patella have been conducted in previous studies; however, the patellar ridge has rarely been reported in the literature. Some authors have indicated that the patellar ridge can act as a fulcrum in the flexion activities of knee joint, thus influencing normal kinematics[12].Using cadavers, Gaurav et al. assessed the position of the patellar ridge from the proportion of the medial facet width within the whole width[17]. They found the patellar lateral and medial facet width ratio to be 1:3; the patella ridge was located in the medial facet. Yoo et al. reported that the central ridge of patella was located 19.9 mm lateral from the medial border by means of magnetic resonance images[18]. Several scholars have investigated the anatomic characteristics of the patellar ridge and the matching relationship between the patellar ridge and the femoral trochlear groove. Their results indicate that the anatomic morphology of the patellar ridge allows the patella to engage with the trochlear groove. If the patellar ridge and femoral trochlear groove had a parallel relationship, that would impede the engagement of the patella into the trochlear groove[19]. Therefore, patellar ridge is very important for the function of patellofemoral joint. Using CT and 3D reconstruction, we also found the patellar ridge to be an oblique line from the proximal to the distal medial patella, located in the medial patellar facet. This special structure of patellar ridge perhaps is of great significance for the diagnosis of patella instability and maltracking.
In TKA with the patella resurfacing, the poor position of the patellar prosthesis can lead to complications such as patellar instability, aseptic loosening, excessive wear, anterior knee pain, or implant failure[20]. How to get proper position of the patellar prosthesis is an important technical issue in TKA, which has been less researched. Several technical factors may contribute to identify the optimal position of patellar prosthesis. Medialization of the patellar prosthesis during patellar resurfacing has been recommended with less need for lateral retinacular release, and significant decrease in patellofemoral contact force[21]. Clinical studies have shown definite decreases in the rate of lateral retinacular release as a result of medialization of the patellar component[22]. With the consensus being in favor of medialization, many surgeons place the patellar component based on the position of the median ridge. Hofmann et al. emphasized that the surgeon should attempt to place the high point of the component at the location of the median ridge, which is the normal high point of the patella in the individual patient[23]. However, R.H. Lee et al. found that native patella vary widely in their axial plane alignments, and this was affected by the various possible positions of the median ridge. According to that data, a more medial placement of the median ridge tended to be associated with worse initial patellar tilt[24]. Increased lateral patellar tilt can cause several problems, including incorrect loading of both the patella and the extensor apparatus, patellar impingement on the prosthesis, and erosion[25]. On the other hand, in some patients such as East Asian patients who had small medial facet, replicating the original center of the patellar ridge the median ridge can lead to overhang of the medial prosthesis. The native anatomy and kinematics of the patellofemoral joint vary by individual. From previous literature, it is unclear which correlations exist between the center of the patellar ridge and the patellar cut after resection in the knee joint. Although many authors advocate the intermediation of patellar components, and for all patients, their methods are usually empirical, such as no more than 2.5mm or lack of quantification, such as the proposal to place buttons in two-thirds of the medial patella[26, 27]. Our study has some news findings, in 97.5% of the patients, the centre of the patellar ridge was found to be medial to the centre of the patellar cut. If the centre of the patellar ridge was regarded as optimal position for patellar prosthesis, the results of this study were inconsistent with those found which reported an optimal position being medial to the centre of the patellar cut in every case. Additionally, placement of the patellar component in relation to the vertical axis of the patellar cut has been even less reported. Lee TQ et al. concluded that a distal placement of the patellar component should lead to decreased loading at higher knee flexion angles[28]. Our results show that the center of the patellar ridge was located more proximally relative to the centre of the patellar cut in 91.25%of cases with the others located more distally. Of course, our data are based on CT-based 3D computer model, further clinical confirmation still needed.
Though the newly developed CT-based 3D computer model could be advantageous to evaluating patellar dimensions, there were also some limitations in this study. One limitation was that the CT scans used to construct the 3D patellae did not include the cartilage thickness, which should be taken into account when preparing for patella resection. The other limitation was that we measured the knees of 20- to 40-year old healthy subjects only. The patella of an arthritic knee might be excavated and reduced substantially by degeneration. Finally, it's worth noting that a major limitation of the current study is that this was a purely anatomical study and had no clinical relevance. We do not know whether changes in patella ridge have significant clinical consequences. Therefore, our interpretation of the data was based on the previously published reports, expert opinion, or basic scientific research, on what constitutes a good surgical technique for patella surface replacement. Despite these limitations, we believe that our study developed a valid and reproducible method for patellar measurement and virtual resection.