Is There an Advantage with Measured Resection for Femoral Rotation Resection in TKA? A Randomized Controlled Trial

Background Total knee arthroplasty (TKA) is performed by gap balance (GB) and measured resection (MR) techniques for femoral rotation resection to achieve knee flexion balance and knee stability. It is still controversial to choose which technique is more advantageous. The purpose of this study was to compare the early clinical efficacy of GB and MR in TKA. This study was a prospective randomized, double-blind controlled trial. From March 2018 to March 2019, 99 patients (99 knees) who underwent TKA at our institution were randomly treated with GB and MR. The cutting thickness of posterior condyle, radiographic findings, medial and lateral compartment pressure, gait analysis, and patients satisfaction were recorded.


Introduction
Total knee arthroplasty (TKA) is one of the effective methods to treat middle-stage and end-stage knee arthritis [1]. According to reports, more than 600,000 people in the United States undergo TKA surgery each year. However, the study shows that 19% of patients are not satisfied with the outcome of TKA [2]. In TKA, the precise placement of the prosthesis position has a great impact on the clinical effect [3,4].
Therefore, it is a key step to obtain a symmetrically flexion and extension gap to match the prosthesis in the success of TKA. Certainly, the femoral extorsion resection directly affects the flexion gap. Appropriate rotation resection can obtain a good flexion gap, which is beneficial to the functional recovery and obtain good contact pressure and patellofemoral joint motion [5,6]. Asymmetric gaps and incorrect soft tissue balance can lead to instability of knee, patellofemoral joint dysfunction, and anterior knee pain [7].
Gap balance (GB) resection and measurement resection (MR) are two techniques of femoral extorsion resection in TKA surgery [8,9]. MR technique was performed by the positioning of the Transepicondylar Axes (TEA), Anterior posterior (AP), and Posterior Condylar Axes (PCA) to obtain the flexion gap [3,[10][11][12]. In the surgical operation, however, it is difficult for the surgeon to accurately identify the bone markers. Yau [13] found that 56%,72% and 60% of femoral rotation exceeded the normal range when resection was performed using the TEA, PCA and AP, respectively. GB technique was proposed by Freeman and Install [14][15][16]. Based on the balance of soft tissue and ligament tension, the distal femoral flexion and extension resection are parallel to the tibial proximal platform. But GB technique may cause internal rotation of a part of the patients' prosthesis and poor patellofemoral joint motion. Therefore, there is still controversy about which surgical technique is used to achieve optimal gap balance.
Whether GB or MR technique, the primary goal is to ensure the stability of the knee prosthesis and good function. Previous studies have compared the two techniques in terms of knee stability, radiographic findings and patient satisfaction [17][18][19], but the results are still controversial. Moreover, most studies only use subjective indexes such as patient satisfaction, pain score or functional score, and cannot fully evaluate the knee joint performance in daily life. Few studies have used objective gait analysis (Vicon MX) to compare postoperative knee function in two techniques [20,21]. Therefore, we designed a prospective, double-blind, randomized controlled study comparing these two techniques. Through postoperative gait analysis, intraoperative resection and pressure index, postoperative radiographic findings, and subjective satisfaction of patients, the clinical effect was evaluated in two techniques. We hypothesized that GB technique has advantages in early postoperative knee stability and knee flexion function improvement.

Randomization and masking
Basing on the above selection and exclusion criteria, we prospectively investigated the clinical data of 110 patients who received primary TKA from March 2018 to March 2019. Krishna [20,22] et al. assessed the clinical effect following GB and MR technique for TKA. Based on their results and considering the potential loss to follow-up of 10%, a total of 110 patients were needed. A computerized randomization list was created by a statistician in a random order using a concealed block size of four. The list was used to randomly assign knees to either the GB or MR group. Allocations were sealed in consecutively numbered opaque envelopes.
Once the patient agrees to be included in the trial, the next sealed envelope is opened by an independent investigator to randomly assign a resection technique.
Patients and clinical data collector were blinded to the group assignment. The data center based at the Department of Orthopedic Surgery was responsible for treatment allocation, central monitoring, and statistical analyses.

Gait analysis
The British 10-Camera Vicon 3D Motion Capture System was used to analyze the postoperative gait of patients ( fig 2). Then Polygon Software was used to analyze the temporospatial and kinematics parameters of the gait cycle. Patients wore tights and marked the lower body with a reflective markers. By Capturing the position of the reflective markers at static state, calculation of the alignment of the knee in Sagittal, Coronal and transversal plane was optimized. Patients should walk barefoot on platform at a self-determined walking speed during 10 captured dynamic trials. Kinematic data was collected by Capture System. Our primary goals were temporospatial and kinematic parameters of the sagittal, conormal and transversal plane. degrees of knee joint movement during swing phase were calculated in the three planes. Knee rotation angle was defined between the plane formed by the femoral reflective markers and the plane formed by the tibia reflective markers in transversal plane (Fig 2). Direction of rotation (positive internal vs. negative external rotation) depended on the position of the moving element compared to the fixed element. At the end of the trials, Polygon Software was used to analyze the temporospatial and kinematics parameters of the gait cycle.

Procedures
All patients received general anesthesia in the supine position. Pneumatic tourniquets were used in the surgery. Routine disinfection was performed and towels were applied. Intraoperative femoral extorsion resection was performed by GP and MR techniques, and the distal femur and proximal tibia resection were performed using the same technique. Both groups used the same prosthesis (Zimmer, Posterior stabilized Prosthesis, PS). The incision was closed and drainage was placed When the knee was bent at 45°. The surgical procedures were performed by the Senior surgeon (X.Y.C). Before this study, Dr. C had independently completed hundreds of TKA with GP and MR techniques, and had extensive clinical experience.
MR: After the distal femur and proximal tibia resection, the knee joint was flexed 90 degrees and femoral external rotation positioning system was selected for external rotation resection. Use the femoral size measurer to determine the femoral prosthesis model, refer to the Posterior Condylar axis (PCA), and select the external rotation angle (usually 3° external rotation) to determine the osteotomy surface.
The femoral size measuring device was used to determine the femoral prosthesis size, and the external rotation angle (3° external rotation) was selected according to PCA to determine the cut thickness. We used the 4-in-1 cutting block localization to perform osteotomy of the anterior condyle, posterior condyle, and intercondylar fossa. A spacer block was placed into the extension gap to confirm it was large enough to accommodate the prosthesis. If the extension gap is unbalanced, the AP and TEA will be used to adjust the external rotation angle for resection or soft tissue release to achieve gap balance. Measurements were recorded by independent researcher.

Post-operative treatment
All patients used the same nutritional interventions, pain management, antibiotics, thromboprophylaxis, and the same care and rehabilitation exercise regimen. The

Statistical analysis
Data were processed with SPSS 23.0 statistical software and expressed as mean values ± standard deviation (x±s). The categorical variables were compared with a chi-square or Fisher's exact tests, and continuous variables were compared with a Student's t-tests. Data among groups were compared with paired t tests. The test level α value is 0.05 on both sides. The difference was statistically significant at P < 0.05.

Results
A total of 110 patients (110 TKAs) enrolled in the study received unilateral THA, and eligible patients will be included in the analysis. Three patients withdrew from the study. Three patients had incorrect resection technique. The data of five patients were lost. This condition resulted in 99 knees in 99 patients for analysis at 12 months follow-up (Fig 1). The surgeon (X.Y.C) completed all operations.  Fig 3).
In pressure measurement, medial compartment pressure was significantly higher than lateral compartment at 90° flexion and full flexion in MR group (10.9±7.1 compared with 7.7±5.3, 9.0±5.2 compared with 7.1±4.7, P<0.05, Table 2). In GB group, there was no statistically significant difference in the medial and lateral pressure. In the process of passive flexion, the medial and lateral compartment pressure decreased with the increase of the degree of flexion. The pressure difference in GB group decreased gradually, while the pressure difference in MR group did not change much (Fig 4).
In gait analysis, Sagittal max knee flexion range was significantly different between the two groups (52.77±5.40 compared with 49.52±7.83, P<0.05, table 3). In the GB group, transversal knee rotation range was significantly larger than that in the MR group (3.60±2.51 compared with 2.47±2.52, P<0.05, Fig 5).

4.Discussion
In posterior condyle resection, the cutting thickness in GB group was significantly higher than that in MR group, which obtained a rectangular gap. Therefore, in the gait analysis, the Sagittal knee flexion and the Transversal knee rotation were significantly higher than those of the MR group. Krishna's study [22] are similar to ours. However, some studies have shown that the thickness of posterior condyle in GB is lower than MR [23]. In addition, the cutting thickness is related to the soft tissue release. The cutting thickness of MR is less than GB, and soft tissue release is needed in the later stage. However, less soft tissue release leads to less trauma, which not only helps to reduce postoperative pain, but also contributes to rapid In the pressure measurement, it can be found that the medial pressure between two groups is significantly higher than the lateral pressure in Fig 3. And the pressure has a decreasing trend as the flexion range increased. Previous studies have shown that the knee joint is nearly stable when the pressure difference of medial and lateral compartment is no more than 15 pounds [28,29]. In this study, the difference was basically within 6 pounds. The conclusions of Kyn et al. are similar to this study [22,30]. Then Tokuhara [31] et al. measured the medial and lateral gaps of the normal knee by MRI technique and it was found that gaps difference was as high as 4.6 mm when the flexion was 90°. This also indirectly reflects the pressure difference of the medial and lateral compartments. In addition, Nagai [32] measured the medial and lateral gaps under different tension conditions in GB resection. As the tension and flexion range increased, the medial and lateral gaps also increased but the gap difference did not change much. The rectangular gap obtained and soft tissue balance by GB are better than MR.
Previous studies of gait analysis showed that walking speed and step length of patients with knee osteoarthritis are significantly reduced compared with normal people, and the single support time in the whole gait cycle is significantly longer.
The above parameters of patients after TKA will be improved compared with the preoperative parameters, but there is still a certain gap compared with normal gait [33][34][35]. This study mainly compared the effects of GB and MR techniques on postoperative gait in TKA. Only sagittal max knee flexion and transversal knee rotation were significantly different between the two techniques (Table 4). Maier's gait analysis of patients after TKA showed a decrease in sagittal max flexion during the swing phase. This conclusion is similar to this study [36]. Huang [37]evaluated the effects of two techniques on the kneeling function after TKA. It showed that GB technique has certain advantages in improving kneeling function, joint mobility and functional improvement after 2 years. One reason may be that the Cut Thickness of posterior condyle in GB was significantly higher than that in MR, which could form a good rectangular flexion gap. So, the movement is closer to normal during flexion.
Our study had several limitations. First, this study only compared the two groups and did not compare with the contralateral knee or the normal knee. In addition, due to the complexity of the gait analysis process and the lack of time for postoperative patients, only 83 patients were available for analysis. Those patients cannot serve as a random sample representative for the entire population of patients, but there is no doubt that this study is also valuable. Second, range of motion was additionally limited by patients pain which can lead to reduced or evading movements. These can affect the collection and analysis of data. Third, the follow-up time of this study was only 12 months. It took longer time to follow up the functional evaluation of the knee. Four, the majority of patients recruited in this study may be different from other regions in knee anatomy [3,6]. The demographic characteristics should be noted before extrapolating to other populations.

5.Conclusion
In TKA, the use of GB or MR techniques for femoral external resection has little effect on daily walking, but the GB technique has advantages in early postoperative knee stability and knee flexion function improvement.

Ethics approval and consent to participate
This study has been approved by the Ethic Committee of the Affiliated Hospital of Xuzhou Medical University (XYFY2018-JS004-01). All patients received informed consent and signed the informed consent.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article, and the supplementary file. We do not wish to share our patients' data because it involves patient's privacy.

Competing interests
The authors declare that they have no competing interests.

Funding
The authors received no financial support for the research, authorship, and publication of this article.

Authors' contributions
YeZ, SF did the study, analyzed the data, and wrote the manuscript. YuZ, JNS, LA,XYC, GCZ was involved in the design, data management, and analysis of the study. All authors read and approved the final manuscript.   Less than differential loading of 15 pounds between the medial and lateral compartments, was considered as adequately "balanced" according to previous studies.  Cutting thickness of medial and lateral compartment.

Figure 4
Pressure changes of medial and lateral compartment.

Figure 5
Sagittal knee flexion during gait cycle.