A Key Technique for Lateral Unicompartmental Knee Arthroplasty and Its Clinical Efficacy


 Background: Currently, the surgical technique and design of surgical instruments for lateral unicompartmental knee arthroplasty (UKA) in treating lateral knee compartment-specific anatomical structures cannot be replicated. Moreover, few articles have reported on the surgical techniques and clinical effects.These have caused most joint surgeons to dare not to carry out this operationinChina.Here, we report on how we perform this procedure in a replicable manner using redesigned tools, and evaluate subsequent American Knee Society scores, range of motion (ROM), and revision rates.Methods: In this retrospective review (January 2014 through January 2016) we identified 158 patients (35 males and 123 females) who had 160 lateral unicompartmental arthroplasties with a minimum of 36 months follow-up (mean 54, range 36–60 months). The initial symptoms of painful osteoarthritis or osteonecrosis were localized to the lateral compartment. The standard of X-slices is mainly limited to the lateral compartment (Ahlbäck grade II OA or worse) while the articular cartilage of the medial compartment shows no obvious clinical or imaging signs of wear. Any valgus deformity must be correctable to reduce varus stress, and any form of inflammatory arthritis is an absolute contra-indication for lateral UKA.Results: At follow-up, American Knee Society scores (0–100 points) averaged 47.3 ± 1.5 for pain, 97.0 ± 1.7 for clinical outcome, and 97.1 ± 4.1 for function, and ROM averaged 125.5 ± 5.3°. None of the patients had reoperations or revisions.Conclusion: These standard surgical techniques and tools are useful to beginners because they can effectively avoid impingement of the femur against the tibial spine eminences and provide good results.


Introduction
Unicompartmental femoro-tibial osteoarthritis (OA) usually affects the medial compartment of the knee, but the lateral compartment is primarily involved in 10% of the cases [1]. Femoral osteotomy is more relevant for young patients with lateral compartment OA and signi cant valgus deformity [2]. However, the restoration of knee joint function after surgery for OA with Ahlbäck grade II and worse has limited e cacy [3,4]. Because of differences in the kinetic mechanisms of the lateral knee compartment, the effect of varus osteotomy is not as satisfactory as that of valgus osteotomy [5]. In addition, because of previous surgical incision and partial bone removal, total knee arthroplasty remains very di cult, and improvements in knee joint function after total knee arthroplasty are not satisfactory [6]. Unicondylar replacement for medial or lateral compartments began in the 1970s. For a patient with singlecompartment OA, unicondylar arthroplasty offers faster recovery and better functional recovery than total knee arthroplasty [7,8]. In addition, it retains a large amount of bone, making it easier to complete knee revision later [9,10]. In the past, most studies considered that the failure rate of lateral unicompartmental knee arthroplasty (UKA) in the treatment of lateral compartment OA was high, reaching up to 17% [11][12][13][14][15]. As a result, some surgeons believed that total knee arthroplasty would be the only effective treatment for lateral compartment OA [16]. However, with recent progress in technology, improvements in implant prostheses and tools, and strict screening of patients, the 10-year survival rates of prosthesis among patients undergoing lateral unicondylar replacement have exceeded 90% [17][18][19][20]. Because of fewer surgical indications and the functional anatomical structure of the lateral compartment, lateral unicompartmental replacement is technically more challenging than medial unicompartmental replacement [21,22]. Here, we review the current surgical indications for lateral unicompartmental replacement, patient selection, surgical techniques and postoperative clinical outcomes.

Study Design and Indications
In this retrospective review (January 2014 through January 2016), we identi ed patients who had undergone lateral unicompartmental arthroplasties. Our indications for lateral UKA are painful isolated osteoarthritis and osteonecrosis. The lateral compartment of the knees showed signi cant loss of joint space on radiographs (Ahlbäck grade II or worse), and the medial compartment of articular cartilage was intact on clinical imaging. Any valgus deformity has to be correctable to reduce stress and no signi cant femoropatellar problems or serious maltracking of patella. All forms of in ammatory arthritis are absolute contra-indications to lateral UKA.

Physical Examination
During the clinical examination of a lateral UKA knee joint, the surgeon must evaluate the mobility of the joint. We require a exion of 100° and extension should be completely normal. If the patient has any pain in the front of the knee, then an assessment of the patellofemoral joint (PFJ) is essential. The stability of the sagittal and coronal joints should be evaluated carefully. Assessment of the anterior cruciate ligament (ACL) requires prior noti cation to the patient. Assessment of knee joint pain and swelling axial displacement tests might be limited. Any valgus deformation should be correctable upon varus stress [23]. Imaging X-ray imaging should include the antero-posterior view, a lateral view, Rosenberg-View or stress radiographs in exion (varus and valgus-stress in exion by hands of the examiner). Ollivier et al. argued that the valgus deformity must be correctable by varus stress [23]. Internal and external stress-bearing Xrays should be taken with the patient in a supine position, and the surgeon should bend the knee when internal and external tumbling stresses are applied by the intraoperative C-arm machine. This view is essential to assess the loss of thickness of the medial knee cartilage and whether the valgus deformity can be correctable to a neutral position. For valgus deformities that cannot be correctable, soft-tissue release is necessary and should be treated with total knee replacement. Lateral radiographs should be used to assess whether the displacement of the tibial plateau of the knee is >10 mm, or whether wear of the tibia is present in the front, middle or rear of the knee to assess the integrity of the ACL. The humeral axial X-ray section is mainly used for con rming the lack of stenosis in the patellofemoral joint space. Occasionally, if the status of the ACL cannot be determined, then magnetic resonance imaging (MRI) or arthroscopy is necessary.

Surgical Technique
The procedure is performed with the patient under spinal or general anaesthesia on a standard operating bed with two leg rests and a blood strap. A stent is placed on the proximal thigh of the affected limb, causing the hip on the affected side to bend to 60°, and the knee joint will sag naturally. One leg rest is placed behind the thigh and the hip is exed to 45°. The target knee is disinfected conventionally, and exion is maintained at least 100°. Since the patella is much more disturbing the lateral preparation compared to medially, the surgeon rst should reduce the lateral edge of the patella. The upper edge of the incision is at on the upper patella, and the distal incision is 2cm below the joint line and the lateral edge of the tibial tuberosity. The joint capsule is cut open and the lateral infra-orbital fat pad is partially removed to facilitate observation of the femoral condyle, the ACL and related lateral tibial platform ( Fig.   1). It is important to emphasize that the ligament balancing principle cannot be applied to lateral UKA. The lateral coronary ligament cannot be balanced or loosened, and the knee exion is maintained at 60°t o observe the tension of the ACL, the medial compartment and the patellofemoral joint (Fig. 2). Next, the intercondylar spurs should be removed to prevent any impact on the ACL. Before completion of vertical osteotomy of the femur, there's a small sulcus on the anterior part of the lateral condyle. However, it is sometimes not obvious. So the front contact points of the femoral condyle and the tibial plateau should be marked (Fig. 3). It should be the anterior border for the femoral component.

Tibial Preparation
According to the anatomy of the lateral compartment, the width of the tibial cut need to be tested by the tibial prosthetic test tool. It has three sizes of 45, 50, 55mm. Sagittal cut line is drawn with the electrocautery between the medial border of the lateral femoral condyle and the lateral intercondylar spine (Fig. 3). The tibial cutting guide is placed ush with the distal of the prosthetic condyle(11mm) edge using a marked point on the front lower side of the tibial plateau. The 11-mm cutting height corresponds to the plateau (9mm), plus a 2-mm laxity safety margin. The saw blade is internally rotated 15 degrees relative to the vertical line of the front edge of the tibial cutting guide to carrying out a sagittal cut through the previously marked points. Horizontal cut is then performed (Fig. 4). The tibial resection should be as small as possible. At the same time, it is necessary to protect the cortex of the anterior and posterior tibia as far as possible, as this can not only important for avoiding a posterior subsidence of the implant but also for preventing a tibial fracture. The reclining angle is generally set according to the tibial cutting guide at usually 0-3 °. For cut on the sagittal plane, a vertical saw applied perpendicular to the leading edge of the tibial cut module is used. This standard procedure is simple and practical. Some surgical teams recommend splitting of the patellofemoral ligament. However, we use retraction of the patellar ligament and osteotomy after exposing the osteotomy. During the operation, the impact of the femoral condyle and intercondylar spine caused by the drop of the joint line after cut of the tibial is observed (Fig. 5).

Femoral Preparation
After tibial cut, the appropriate dimensions of the test models are selected according to the test results. The appropriate test models are then placed into the lateral space to restore the normal force line of the knee joint. Then parallel incisions on the test model of the tibia are cut using a swing saw to remove the sclerotic bone of the subchondral bone. This not only avoids the valgus and varus of femoral osteotomy, but also prevents excessive deviation or deviation of the prosthesis (Fig. 6). The midline of the femoral condyle prosthesis is the femoral condyle that corresponds to the midline marking line on the tibial plateau test model (Fig. 7). First, the femoral condyle is marked using an electric knife. The tibial test model is removed and then the knee joint is exed extremely. The femoral drilling test model and special limit device are used for drilling tests. When drilling, the midline of the femoral model must coincide with the marker line on the femoral condyle, and the drill is placed at about 30° to the posterior cortex of the femur (Fig. 8). Particular attention should be paid to avoid too large a femoral prosthesis. We recommend that the depth of cut should be 1-2mm below the cartilage to avoid a step between the femoral prosthesis and the patella. The position of the femur drilling guide should be not in internalrotation or outerrotation, both for avoiding impingement to the tibial spine eminence in extension and also for minimizing contact to the patella. Once the femoral condyle cut is completed, the cutting module is removed. The removal of osteophytes in the posterior femoral condyle and soft tissues occupying the posterior space is necessary to improve knee exion and prevent the impact of the polyethylene prosthesis in high exion positions.

Prosthesis Selection and Placement
Next, the type of tibial gasket should be chosen according to the maximum coverage area of the tibial plateau. Theoretically, there should be no need for anterior, posterior or lateral prosthesis drapes. In practice, there should be at least no lateral or anterior edge drapes. The femoral prosthesis should match the lateral femoral condyle as closely as possible. If two adjacent models are acceptable, then the smaller one is recommended. First, the femoral prosthesis is placed, followed by the tibial prosthesis. At the same time, the tibial plateau is moved forward as far as possible to prepare for the smooth placement of the tibial prosthesis. The exion-extension gap is tested by placing the test die and lining. Because of the different anatomical structures of the medial and lateral compartments, the joint space of the lateral compartment should be about 2mm larger than that of the extensor when exing the knee at 90° (Fig. 9).
At this stage, whether the femoral side impacts on the tibial intercondylar spine during knee extension should be checked carefully. We have not found any impacts between the femoral prosthesis and the tibial intercondylar spine during knee extension and exion tests. We recommended that the midline of the femoral prosthesis should be aligned with the midline of the tibial prosthesis. When the knee is bent at 90°, the median surfaces of the femoral and tibial prosthesis are aligned relatively, achieving good surgical results (Fig. 10). The valgus deformity of lateral UKA cannot be overcorrected to prevent the pressure of the medial compartment from increasing, which guarantees good long-term surgical results.
When placing the prosthesis, we take the following steps. First, the knee joint is exed extremely and rotated to increase the exposure of the lateral compartment. The femoral cement prosthesis is placed, the cement around the prosthesis is removed and then the tibial cement prosthesis is placed. Once the tibial prosthesis has been inserted, the knee joint appears nearly straight, and the bone cement behind the tibial plateau prosthesis is removed with a matching arc curette. Then the knee is bent at 45° and a 2-mm insert is inserted into the gap. After removing the excess cement and drying the bone cement, the insert is removed.

The Screw Home Mechanism
During the nal stage of knee extension-from a knee exion of 20° to knee extension-the tibia is rotated outwards, tightening the two cruciate ligaments and locking the knee joint. Because the tibia has greatest stability relative to the femur, some studies recommend that the choice of femoral prosthesis position is very important, as otherwise knee extension can lead to excessive internal rotation, impacting on the tibial intercondylar spine. Therefore, when bending the knee, the femoral prosthesis should be rotated as far as possible and laterally [23]. Preoperative imaging data and intraoperative observations have shown that the impact between the medial edge of the lateral femoral condyle and the tibial intercondylar spine occurs only in patients with narrow lateral compartments. After the height of the lateral articular surface has been restored during the operation, the impact disappears. Therefore, we believe that this impact may be mainly related to whether the joint line is restored. The width of the normal intercondylar concave is su cient to avoid the impact between the femoral condyle and the tibial intercondylar spine during locking.

Discussion
Obesity is not considered a contra-indication for lateral UKA. Berend et al. reported that excess weight can worsen prosthesis wear [11]. However, other studies indicate that there is no speci c relationship between BMI and surgical outcomes. Rather, the wear of a unicondylar prosthesis is not related to body weight but to the amount of exercise [23]. Patellofemoral arthritis and periarticular osteophytes are also not regarded as contra-indications for UKA, but severe damages laterally in the PFJ with associated symptoms should be a relative contraindication. With correction of the valgus, the lateral release due to the lateral approach and the partial resection of the lateral facette the tracking of the patella is automatically improved, but severe maltracking can not be corrected. Any type of in ammatory arthritis is an absolute contraindication for lateral UKA. Therefore, it has been suggested that as long as the patient has bone-to-bone arthritis, the ACL will remain intact, so the patient's age, weight and early onset of patellofemoral arthritis are no longer issues that need to be considered [16,23].
The use of X-rays is mainly con ned to the lateral compartment for OA of Ahlbäck grade II or worse when the medial compartment articular cartilage has no obvious clinical or imaging signs of wear and tear. A Roseberg view is taken at 45° of knee exion, when the X-ray tube is placed behind the knee joint parallel to the articular line. According to Kozin and Scott, the valgus deformity before UKA should be <15°. Ollivier et al. [23] reported that the valgus deformity can be completely correctable by varus stress. For those deformities that cannot be correctable, lateral UKA operations cannot be considered because of the presence of subluxation on the coronal plane. Because lateral UKA is mainly used for surface replacement, it might not correct for severe bone loss or OCN [16,23].
Several studies have con rmed good long-and medium-term effects for medial UKA. The 10-year prosthesis survival rate in one study was 95% [24]; however, there are limited data on the long-term followup of lateral UKA [25]. It has been suggested that the clinical e cacy of lateral UKA is inferior to that of medial UKA. The dislocation rate of a lateral mobile gasket for UKA was reported to be 10%, and the 5year prosthesis failure rate 21% [13]. Indeed, femoral rollback during knee exion is important for lateral knee compartment functioning. This also partly explains the reasons for the early failure of UKA using mobile gaskets. Scott et al. followed up 12 cases of lateral UKA for 3.5 years and found that six failed [26]. Similarly, 20 cases of lateral UKA were followed up by Cameron et al. and nine of these had failed. The failures were blamed on osteoporosis of the lateral tibial plateau [12]. Insall and Laskin indicated that as long as patient choice was appropriate, lateral UKA achieved good clinical outcomes. Argenson et al. also noted that lateral UKA was an ideal choice for the treatment of lateral compartment OA [26]. Studies have shown that lateral UKA has similar clinical effects to medial UKA [20,27] A retrospective study by Ernstbrunner et al reported that the main causes of lateral UKA failure included OA progression (30%) and aseptic loosening (22%), and these were followed by less common causes of failure including instability (7%), unexplained pain (5%), infection (5%), polyethylene gasket wear (5%) and gasket dislocation (5%). Gasket dislocation was the most common cause of early failure (29%) and also the most common cause of overall failure (27%) for patients with a mobile gasket prosthesis. Among the mid-and late-stage failure cases, OA progression led to the highest failure rates (59% and 78%, respectively). Gasket dislocation is also the most common cause of failure of xed gasket prosthesis (44%) [31]. In order to avoid complications as much as possible, we recommend that the following guidelines should be heeded: (1) strict compliance with the above surgical indications to screen patients, avoiding the possible complications caused by improper indications; (2) use of a xed gasket prosthesis, thus effectively avoiding the occurrence of gasket dislocation; (3) Overcorrection should be avoided; and (4) use of standardized surgical methods so that the operation can be replicated, thus ensuring the reliability and consistency of the operation's effect.

Conclusion
We believe that: 1. The intraoperative method of tibial cut using the method of 15 degrees of internal rotation is di cult to be accurately implemented. Our method of using the saw blade is internally rotated 15 degrees relative to the vertical line of the front edge of the tibial cutting guide is intuitive and accurate. 2. Whether the femoral prosthesis will collide or not with the sacral spine is mainly related to the restore of the normal height of the lateral tibia. In order to accommodate a more uniform distribution on the tibial prosthesis, we chose a method of aligning the midline of the femoral prosthesis with the midline of the tibial prosthesis. 3.To prevent the internal and external rotation of the femeral osteotomy, we used the femoral cut parallel to the tibial plateau. Our surgical technique uses a more intuitive reference, avoiding ambiguous subjective experience. It is easy to learn for beginners who meet often the problems where the prosthesis loosening or impact between the femoral condyle and the iliac crest. And it can also guarantee a good operation result. Follow-up (months) 54 Notes: Numerical expressions are means ± standard deviations.