Femoral Osteotomy For The Rotational Alignment Of The Femoral Components In Total Knee Arthroplasty For Osteoarthritis Patients With Varus Deformity

BackgroundWith the arrival of the aging society in China, the number of patients with knee pain and dysfunction caused by joint degeneration keeps increasing. The treatments of knee osteoarthritis are various.In the guidelines of the treatments for hip and knee osteoarthritis released by Osteoarthritis Research Society International, 25 therapeutic methods were recommended, including conventional treatment, non-drug therapy, drug therapy, surgical treatment and so on. Surgical treatments include unicompartmental arthroplasty, osteotomy, arthroscopy, palliative knee arthrodesis, and total knee arthroplasty(TKA), of which the TKA has been considered as the gold standard and become the most popular. However, the effect of TKA is depended on multiple factors, such as the mechanical and rotational alignment of the lower limbs and the soft tissue balance. The rotational alignment of femoral components is the key factor which affects the function and lifespan of the knee prosthesis in post-TKA knees.There are many methods to determine the orientation of femoral osteotomy for the femoral rotational alignment.The measured resection(MR) and balanced gap(BG) are two major techniques used in clinic. Because the learning curve of the MR technique is relatively flat, it has become the most popular method adopted by surgeons. For the MR technique, the orientation of rotational osteotomy is determined based on a few bony landmarks of the distal femur during the preoperative and intraoperative measurements. The commonly-used reference axes include the surgical transepicondylar axis(STEA), the clinical transepicondylar axis(CTEA), the anterior posterior line(APL), and the posterior femoral condylar line(PCL). Many studies have shown that the STEA was parallel to the flexion-extension axis of the knee, thus it has been considered as the most reliable anatomical reference axis. However, the medial sulcus of the femoral epicondyle is vulnerable to knee degeneration and is likely to be covered by the medial collateral ligaments, making its position difficult to be localized interoperatively. To solve this problem, many surgeons employ the APL, CTEA or 3° externally rotated to the PCL(3° PCL) to align the femoral components, which may lead to uncertain outcomes. The function and longevity of the TKA are highly dependent on the correct alignment of the femoral components. In the present study, we measured and compared multiple anatomical axes with the STEA, aiming to find an appropriate alternative to the latter for the femoral rotational alignment.Currently, the majority of osteotomy plates used for distal femur were designed according to 3° PCL. However, this feature was based on the anthropometry of Caucasian population; one can expect that the anatomy may be different in Asian population. Meanwhile, wear and degeneration of the articular cartilage of the posterior femoral condyles may also change the posterior condylar angle(PCA). Therefore, 3° PCL may not be the optimal method to conduct femoral osteotomy for Chinese population. Some researchers used computed tomography(CT) and magnetic resonance imaging(MRI) to conduct patient-specific osteotomy, which has exhibited good efficacy. Nevertheless, insufficient evidence can be found in the literature revealing the difference between the patient-specific osteotomy and osteotomy referring to 3° PCL. In addition, when evaluating the positions of the prosthesis, traditional CT and MRI can be essentially disturbed by the metal artifacts. Dual source CT(DSCT) is able to eliminate the metal artifacts, facilitating the evaluation of the prosthesis position. In the present study, we conducted a prospective study comparing the two methods for rotational femoral osteotomy: 3°PCL and patient-specific osteotomy. The range of motion(ROM) of the knee under weight-bearing condition, the KSS score as well as the VAS score were quantified. We further used the DSCT to evaluate the accuracy of the prosthesis placement, seeking the theoretical support for the femoral osteotomy.Objectives1. The PCA is altered with the development of knee degeneration, which may lead to inaccurate osteotomy when referring to 3° PCL. Thus using the MRI to quantify the PCA in OA population is necessary. Since the STEA is difficult to be identified interoperatively, surgeons usually use other anatomical axes to estimate its position, which may induce error to the rotational femoral osteotomy. We used the MRI to investigate the relationship between the STEA and other reference axes, aiming to find the theoretical support for the optimal axis as a reference for the femoral rotational alignment.2. To compare the short-term effect of osteotomy referring to 3° PCL and the patient-specific osteotomy based on preoperative CT measurements. The prosthesis placement was also assessed postoperatively using DSCT.Methods1. 86 knees in 86 patients which underwent primary TKA in the center for joint surgery in Southwest hospital from June 2013 to January 2014 were retrospectively studied. Utilizing the PACS system, the STEA, CTEA, perpendicular of the APL, 3° PCL were marked on the axial MRI images. The relationship between the STEA and the other three axes were quantified.2. A prospective study was performed on 41 patients(41 knees) in the center for joint surgery of Southwest Hospital from May 2014 to August 2014. These patients were diagnosed with end-stage OA with varus deformity in the knee. The patients were randomly divided into two groups, 20 knees in the conventional group(3° PCL group) and 21 knees in patient-specific osteotomy group. The weight-bearing knee ROM, Knee Society Score(KSS) and the Visual Analogue Score(VAS) were evaluated postoperatively in the two groups.Results1. In the 86 knees, the PCA was 2.38° ± 1.82°(range-2.06° to 7.06°), lower than the normal value. Wear and degeneration of the articular cartilage were found in all involved knees, including 3 severe cases. The PCL was in external rotation relative to the STEA, with the greatest deviation of 2.06°. 13 PCAs were found to be 0°(15.1%), indicating that the PCL and STEA were parallel; 70 PCAs were found greater than 0°(81.4%), with the greatest value of 7.06°. The CSA(angle intersected by the STEA and CTEA)was 3.78°±0.75°(range 2.04° to 5.67°), approximating 4°. In traditional concept, the APL was perpendicular to the STEA and CTEA; however, this relationship was only found in less than 1/3 cases. The ASA was found to be 5.37° ± 3.17°; the relatively large standard deviation indicated that the APL may not be a reliable reference in patients with varus knee. One-way ANOVA suggested significant differences between angle A and CSA, angle A and ASA, CSA and ASA, respectively(P < 0.05). The average values of the angle A, CSA and ASA were 0.63 °, 3.78 ° and 5.37 °, respectively. These findings suggested that the accuracy of these axes could be concluded as: 3 ° PCL > CTEA > the perpendicular of APL.2. All patients were followed up for 6 weeks. No complications such as infection, loosening of the prosthesis, deep venous thrombosis et al was observed. The weightbearing knee ROM was quantified in all knees. In the traditional group: flexion 100.6° ± 16.0°(range 66° to 128°), extension lag 6.4° ± 6.7°(range 0° to 20°); in the patientspecific osteotomy group: flexion 109.1° ± 16.7°(range 76° to 140°), extension lag 1.5° ± 3.2°(range 0° to 10°). The degrees of flexion in the patient-specific group was on average 9° greater than that in the traditional group, but the difference was not significant(P > 0.05); The extension lag in the patient-specific group was significantly less than the traditional group(P = 0.007). The KSS score in the traditional group was 120.4 ± 22.2(range 86 to 159), of which the clinical score was 65 ± 11(range 48 to 83), and the functional score was 55.3 ± 12.2(range 35 to 75); the KSS score in the patientspecific osteotomy group was 129.9 ± 24.4(range 98 to 179), of which the clinical score was 75 ± 17(range 45 to 99), and the function score was 54.8 ± 13.1(range 25 to 90). The clinical score in the patient-specific osteotomy group was superior to the traditional group with significance(P=0.036), but no significant difference was found in the KSS score and function score(P>0.05). The VAS score was 3.5 ± 0.9(range 3 to 6) in the traditional group and 3.3 ± 0.7(range 3 to 5) in the patient-specific osteotomy group; no significant inter-group difference was found(P > 0.05). The PCA was 0.74° ± 0.96°(range-3.2° to 2.3°) in the patient-specific osteotomy group, and 1.62° ± 1.55°(range-5.9° to 4.6°) in the traditional group, the inter-group difference was significant(P = 0.03).Conclusions1. For patients with end-stage OA as well as varus deformity in the knee, the PCA was lower than its normal value. The PCL was internally rotated in relation to the STEA in a majority of cases, with notable variability. During operations,the 3° PCL can be a more reliable reference than the CTEA and APL. However, attention should be paid to the potential excessive external rotation of the femoral components.2.The short-term function of the TKA knee with patient-specific osteotomy was significant better than that of the traditional group. Patient-specific osteotomy can improve the accuracy of osteotomy, promote the functional movements of the knee, as well as reduce the chance of soft-tissue release.