Early Patellofemoral Function of Unilateral TKA with Medial Pivot Prostheses is Superior than with Conventional Posterior-Stabilized Prostheses

Purpose The purpose of this study was to provide a systematic evaluation of the patellofemoral joint design of Medial Pivot Prosthesis, which incorporates a variety of “Patella-friendly” design features, by comparing clinical and radiographic results with another prosthesis. Methods Early clinical and radiographic results of patients who underwent unilateral TKA with Medial Pivot Prosthesis (The study group, including 126 cases) and conventional Posterior-Stabilized Prosthesis (The control group) were retrospectively compared. Postoperative complications, including anterior knee pain, maltracking, patellar clunk or crepitus (PCC), were evaluated. The postoperative Kujala score and its improvement from baseline in the study were significantly higher than those in the group(Group B). The range of motion (ROM) in group A, including the improvement in ROM, was significantly inferior to group B. In the 90-degree Merchant view, the patellar tilt in group A was smaller than that in group B. Two cases of PCC and 3 cases of anterior knee pain were noted in group A, and 9 cases and 6 cases, respectively, were observed in group B. The incidence of PCC was significantly lower in group A. There were no significant between-group differences in the patella tilt angle at 30 or 60° or in the postoperative patellar translation at 30, 60 or 90°. No between-group difference in posterior condyles angle (PCA) was observed. The KSS scores and WOMAC scores between the two groups were similar. patellofemoral performance attributed to its design characteristics compared to the conventional posterior-stabilized prosthesis.


Introduction
Historically, patellofemoral complications have accounted for 50% of revision surgeries [1]. With the development of contemporary designs and improvements in surgical techniques, complication rates have decreased but remain the most challenging problem after TKA [2]. The incompatible patellofemoral joint design may result in multiple postoperative complications, including patellar anterior knee pain, maltracking, patellar clunk syndrome (PCC) and avascular necrosis [3]. Several in vitro studies have found that patellofemoral kinematics were altered after TKA and that patellofemoral joint pressure was increased compared with the natural knee [4,5]. The use of a knee prosthesis capable of reconstructing normal patellofemoral joint movement and with low patellofemoral joint pressure might be beneficial in reducing postoperative patellofemoral joint complications [6].
The medial pivot femoral prosthesis incorporates a variety of design features that facilitate the reconstruction of natural patellofemoral joint relationships. The femoral prosthesis design features that affect patella function include the shape and depth of the trochlear groove, the thickness of the anterolateral condylar, and the sagittal curve[7-10]. "Patella-friendly" design features of the medial pivot femoral component include a sagittal plane with a single radius of curvature, extending posteriorly and deepening to the natural depth of the trochlear groove, and an elevated lateral edge. Based on these theoretical advantages, the design 5 features of the medial pivot prosthesis may decrease the risk of patellofemoral joint problems [11,12]. Several studies have reported clinical satisfaction and survival analysis of TKA for this prosthesis [13][14][15][16], however, the comparison of clinical and radiographic results of this prosthetic patellofemoral joint with another prosthesis was rare [11].
The total condylar knee prosthesis, based on the concept of the "four-bar link model", was designed in 1973. This classic knee prosthesis is widely used in the treatment of knee osteoarthritis due to excellent clinical and radiological results after surgery [17]. The prototype design used today is considered the gold standard for clinical outcome evaluation after TKA [18]. The conventional posterior stabilized knee prosthesis (cruciate-substituting), one of the classic knee prostheses, is currently widely used in clinical applications. The NexGen LPS-Flex prosthesis is the most representative of this kind of prosthesis. The variable sagittal curvature of the femoral prosthesis results in the motion curve being "frame-like", increasing the risk of poor patella tracking and intraoperatively releasing the This study retrospectively compared the clinical and radiographic results of TKA using the medial pivot prosthesis with conventional posterior-stabilized prosthesis by a matched pair analysis. The hypothesis of the study was that early clinical and radiological results of the medial pivot prosthesis would be comparable or better than those of conventional posterior-stabilized prosthesis, especially in aspects related to patellofemoral function.

Methods
After obtaining the approval of the institutional review board of our hospital and of our patient, all consecutive patients who underwent a unilateral TKA with a medial .7-36.9 kg/m2). There were no significant differences in the demographics or clinical characteristics between the two groups. All TKAs were implanted by the same surgeon (YZW) who had performed more than 500 cases annually. The surgical principles and postoperative rehabilitation protocol were similar between groups A and B. Briefly, under tourniquet control, knees were exposed through a midvastus approach. Osteotomy was performed by measurement.
The femoral prosthesis was implanted at 3° external rotation relative to the posterior condylar axis and 5° valgus relative to the coronal femoral mechanical axis. The rotation of the tibial prosthesis was aligned with reference to the medial one-third of the tibial tubercle. The proximal tibial osteotomy was located 10 mm below the highest point of the articular cartilage on the lateral tibial plateau, perpendicular to the long axis of the tibial coronal plane, with a 3° posterior slope in the sagittal plane. All patellae were unresurfaced and de-nerved with electrocautery. No lateral retinacular release was performed. Patients initiated passive ROM exercise with CPM machine and partial weight-bearing walking training on postoperative day 2.
The clinical results were assessed at the preoperative and final follow-up periods according to the Knee Society's KSS and the WOMAC Index. Clinical results related to patellofemoral joint symptoms were evaluated using the Kujala scoring system, which is widely used to assess subjective symptoms and functional limitations in patellofemoral disorders [20]. At pre-operation and at the last follow-up, Merchant views were taken with the knee flexion at 30, 60, and 90° to measure patella shift and tilt ( Fig. 1)[21]. Patella tilt was formed by the angle between the transverse axis of the patella and the anterior intercondylar line. A positive value of patella tilt indicated that the transverse axis of the patella was tilted outward relative to the anterior intercondylar line [21]. Patella shift was defined as the distance between the intercondylar sulcus and the median ridge of the patella. When the median ridge 8 of the patella was on the lateral side relative to the intercondylar sulcus, we define the shift as a positive value; otherwise, it was considered a negative value [21].
Since the measurement of the patella tilt was based on the anterior intercondylar line, the femoral prosthesis rotation was the main factor influencing the position of the anterior intercondylar line. In addition, rotational deviation of the femoral prosthesis was one of the factors affecting the function of and complications associated with the patellofemoral joint [22]. It is necessary to determine whether there was a significant difference in the external rotation between the two groups.
The measurement of the posterior condylar angle (PCA) was taken at pre-operation and at the last follow-up (Fig. 2). A standard GE Medical Systems CT scanner was used to evaluate the PCA of each knee. The fully extended knee joint was scanned from the distal metaphysis to the tibial tubercle.
To reduce measurement bias, radiographic results were obtained and analysed by the same independent orthopaedic surgeons preoperatively and at the final followup. Imaging results were measured three times, then averaged to obtain the final measurements. Images were read on a PACS General Electric, Chicago, IL, USA monitor and measured with a mouse pointer and automatic computer calculations.
The clinical scores, ROM and radiographic measurements at the last follow-up were compared (Student's t test). The preoperative/postoperative improvement of the above indicators between the two groups was also compared. The difference in the incidence of postoperative complications between the two groups was compared (χ2 test). SPSS (IBM Corporation, USA) version 20.0 was used for the statistical analysis. P <0.05 was defined as statistically significant.

Clinical Results
Postoperative clinical results and changes in the results are summarized in Table 2.
No statistically significant difference was identified in the KSS total score, including knee score and function score, or in the WOMAC score between the two groups after the operation. No statistically significant difference was identified in the preoperative/postoperative improvements of the above clinical scores. However, we found statistically significant differences in the postoperative Kujala scores and the ROMs between the two groups. The mean Kujala score in group A was better than in group B, while the ROM in group B was significantly higher than in group A.
Simultaneously, the preoperative/postoperative Kujala score improvement in group A was observed to be significantly larger than in group B, but the average ROM improvement was significantly different in favour of group B.

Radiographic Results
The imaging measurement data are summarized in Table 3. Postoperative PCA did not show significant differences between the two groups. No significant betweengroup difference was observed in the Merchant view at 30 or 60° of patella tilt; however, at the 90° position, the mean patella tilt of group A was statistically smaller than that of group B. There were no statistically significant differences in the patella shift between the two groups at 30, 60 and 90°.

Complications
At the last follow-up, 3 cases of anterior knee pain and 2 cases of PCC in group A and 6 cases and 9 cases, respectively, in group B were observed. Compared with group B, the incidence of PCC was significantly lower in group A. There were no periprosthetic infections, loosenings, or incidences of patellar maltracking.

Discussion
This study illustrates that the medial pivot prosthesis could achieve satisfactory early clinical outcomes with superior patellofemoral performance compared to the conventional posterior-stabilized prosthesis. In the early development of total knee prostheses, the tibiofemoral joint design was considered most important, with little attention paid to the design of the patellofemoral joint. Cases of revision due to patellofemoral joint complications had been as high as 50% of the total revision rate [1]. With the continuous improvement in and development of prosthetic design and surgical techniques, there are significantly fewer patellofemoral joint complications than before; however, patellofemoral joint problems are still a common complication after TKA. Compared with the natural knee joint, the kinematics of the patellofemoral joint are changed and the pressure of the patellofemoral joint are increased after TKA[4, 5, 28]. It has been reported that using a knee prosthesis that could reconstruct the natural patellofemoral joint movement and achieve low patellofemoral pressure would be beneficial in improving patellofemoral function [6]. Despite claims of a theoretical advantage in the "patella-friendly" design characteristics of the medial pivot prosthesis, no previous prospective clinical study focused on comparing the patellofemoral joint of this prosthesis with other total knee prostheses.
The femoral component of the MP prosthesis incorporates a variety of design features that facilitate the reconstruction of natural patellofemoral joint relationships. The femoral component has a single radius of curvature in the sagittal plane from fully extended to 90 degrees, which is closer to the anatomy of the natural knee [29]. The depth of the natural trochlear groove is restored, so "overfilling" could be effectively avoided in front of the knee, which would be beneficial for extension devices to function normally [7,9]. The anterior lateral edge is 3 mm-4 mm higher than the bottom of the trochlear groove, which is an essential feature for maintaining the patella track in early knee flexion [30]. In addition, the trochlear groove extends backwards so that the patella could also fully contact the femoral prosthesis in deep knee flexion [31].
In this study, the Kujala score, which is widely used to assess functional limitations and subjective symptoms in patellofemoral disorders and TKAs, was 77.16±3.80 in group A, which was significantly better than the 75.97±4.06 in group B. We also evaluated differences in the postoperative improvement of the Kujala scores from the preoperative baseline, with group A predominating. In the 90-degree Merchant view, the patella tilt in group A was smaller than in group B. It has been reported that the incidence of patellar clunk increases by 1.27 for each degree raise in patellar tilt [32]. We believed that group A was also superior to group B in terms of imaging performance. The advantage was attributed to the use of a prosthesis with "patella-friendly" properties and the characteristics of the reconstructed natural patellofemoral joint.  PCA posterior condyles angle P < 0.05 was defined as statistically significant  Figure 1 Patella tilt was defined as angle between the transverse axis of the patella (P) and the anter 26 Figure 2 Posterior condyles angle (PCA). PCA was defined as the angle between the transepicondylar a