Introduction

The treatment of varus tibiofemoral malignment and associated medial compartment osteoarthritis of the knee by a high tibial osteotomy (HTO) is a well accepted treatment option [2, 12, 13, 18, 20, 53, 55, 56, 64]. Lateralization of the loading axis decompresses the overloaded medial compartment of the knee, thereby minimizing pain and inhibiting osteoarthritic degeneration. But although valgus correction is clinically successful, patella infera is a common postoperative complication linked to this technique, with incidence reported as high as 89% [12, 18, 23, 36, 53, 56, 64]. The alteration of patella height in the lateral closing HTO (LCHTO) is caused by interstitial scarring of the patella ligament and bone formation behind the patellar ligament [7, 53], whereas in the medial opening wedge HTO (OWHTO) it is caused by a distalisation of the tibial tuberosity [18, 53, 56]. The clinical sequelae of patella infera generally present as anterior knee pain, patellar locking, crepitus, and reduced range of motion, but may also complicate total knee replacement [36, 49, 64]. Patients presenting with patella infera after HTO or TKA experience greater postoperative anterior knee pain [49, 63]; a complication often attributed to patella malalignment and maltracking [29, 30, 64] and elevated cartilage pressure [8, 10, 17, 32, 47, 58, 61]. Although many surgeons still adopt the LCHTO technique, and it has been suggested that this technique exhibits less patella infera that conventional OWHTO [7], the closing wedge technique is consistently compromised by medial collateral ligament instability and other complications postoperatively [26, 40].

A new technique for preserving the insertion of the patellar ligament and the anatomical location of the tibial tuberosity has recently been described [18]. Unlike traditional OWHTO where a single plane osteotomy is performed proximal to the tuberosity (PTO), the distal tuberosity osteotomy (DTO) OWHTO procedure allows the tuberosity to remain attached to the proximal region of the tibia (Fig. 1). Opening of the wedge does not alter the position of patella attachment at the tuberosity, therefore preventing patella infera complication. Whilst patella height does not seem to be affected, an alteration of pressure in the patellofemoral joint through medio-lateral changes of tuberosity position is an untested phenomenon. Indeed, it has been shown that lateralization of the tibial tuberosity increases the Q angle and lateral patellofemoral contact pressure [22, 28]. Hence, whilst DTO OWHTO may be clinically useful in preventing patella infera, attention must be paid to maintain normal joint kinematics.

Fig. 1
figure 1

Schematic models of a medial opening wedge high tibial osteotomy with distal tuberosity osteotomy, b medial opening wedge high tibial osteotomy with proximal tuberosity osteotomy, c lateral closing wedge high tibial osteotomy

Full size image

Patellofemoral cartilage pressure should be unchanged after knee surgery in order to halt joint surface deterioration. To this end, the changes in patellofemoral pressure following HTO have not been well investigated. Hence, we report on the effect of three different types of HTO on contact pressure within the patellofemoral joint.

Materials and methods

Cadaveric knees

Nine fresh-frozen human male cadaveric knees of (median age 72 ± 8.1; six female and three male) were used. Approximately 20 cm of proximal femur and distal tibia were preserved. All knees had not been operated on previously and had relatively normal tibiofemoral alignment. Biplanar radiographs were taken to exclude any osseous pathology, pre-existing disease, or trauma. The specimens were wrapped in saline soaked cloth, and stored at −30°C in tightly sealed plastic bags. Joints were thawed over 24 h before testing. All testing was conducted under regulations outlined by the University of Western Australia Research Ethics Committee.

Osteotomies

The knees were dissected of all skin and subcutaneous tissue. Proximal medial opening (PTO OWHTO), distal medial OWHTO (DTO OWHTO), and lateral closing HTO (CWHTO) were performed on the cadaveric knees sequentially in the aforementioned order (Fig. 1). (1) DTO OWHTO: The osteotomy was performed as detailed by Gaasbeck et al. [18], so that the tuberosity remained proximally attached to the tibia; thereby maintaining patellar height after opening the osteotomy. Primarily, the transverse tibial osteotomy leaves at least 1 cm of the tuberosity thickness intact, before the osteotomy is continued posteriorly through the tibia. Then, from the tibial osteotomy, the tuberosity was cut distally in the frontal plane towards the anterior tibial cortex. Following medial plate fixation with a 3.5 mm LCP, the distal part of the tuberosity is fixed with a bicortical screw to the tibia to increase the fixation strength. (2) PTO OWHTO: While the medial plate fixation remained in place, the transverse osteotomy of the tibia was completed towards the anterior cortex creating a mobile tuberosity. After removing the screw, the tuberosity was distalised and lateralized to its original position and fixed to the tibia with two screws. For both the medial proximal and medial distal OWHTO, the base of the osteotomy was opened and a hard plastic wedge with a base of 14 mm was inserted (Fig. 2). (3) CWHTO: The medial locking plate was removed while the tuberosity remained fixed with screws to the tibia. A lateral bony wedge with a base of 14 mm was resected. The cortex of the medial tibia was aligned and the lateral gap was closed. The medial side was stabilized with a stapler while the lateral osteotomy was fixed with a 3.5 mm LCP.

Fig. 2
figure 2

Medial opening wedge high tibial osteotomy with distal tuberosity osteotomy

Full size image

Knee simulator (Fig. 3)

This rig was designed to permit free rotation of the tibia during flexion. Each knee was therefore mounted in a customized passive motion rig with the femur potted in a cubic metal cylinder using dental cement (Glastone 3000, Die Stone, Type V, Densply, Perth, AU). The cylinder was lined up with the transepicondylar knee axis. An aluminium ring attached to the tibia articulated with a low-friction roller, thus avoiding the constraints associated with gripping the tibia by hand. A clamp was secured to the quadriceps tendon 2 cm proximal to the superior patella. The clamp was connected to a weight. An 87 N extensor muscle loading force was applied. This load is similar to Elias’ et al. [15]. An additional force of 8.5 N was added to the ITB, and 21 N to the biceps and the semimembranosus/semitendinosus tendon, respectively. This distribution of the forces is similar to the force ratios used by other authors [3, 19, 51]. Each knee was flexed with the quadriceps aligned with the femur (Fig. 3.)

Fig. 3
figure 3

Knee simulation rig

Full size image

Contact pressure measurement

The Pedar™ pressure measurement system was used to assess patellofemoral pressure. The pressure sensor was inserted into the knee joint through a small suprapatellar arthrotomy and placed between the patella and the femoral condyles in the patella groove. The sensor system contains over 80 sensors with a size of 1 cm2each. The sensors have a pressure range of 30–1,200 kPa, with a resolution of 5 kPa. The peak value was taken for each flexion angle. To ensure measurement reliability, each test was conducted four times by taking out the sensor and replacing it in the patellofemoral joint. Hence, each knee value represented the mean of four repeated measurements. The patellofemoral pressure was tested in a total of nine cadaveric knees. The outcome pressures for each treatment group are presented as a mean pressure of all nine knees relative to the mean pressure of the intact knees at 15° flexion.

Statistical analysis

The pressure outcomes of each treatment were presented as relative mean pressure ± standard error of mean (SEM) of the nine knees compared to the mean pressure of the intact knees at 15° flexion, which were designated a value of 1. The student’s t test was used to compare the three osteotomies against the intact knees at each flexion angle, with significance determined by P values less than 0.05.

Results

Patellofemoral contact pressure

Contact pressure in the patellofemoral joint space was analysed at 15°, 30°, 60°, 90°, and 120° knee flexion angles following sequential distal medial opening, proximal medial opening, and lateral closing wedge high tibial osteotomy in cadaveric knees (Fig. 4). The relative cartilage pressure of the intact knees started at 1.0 ± 0.09 at 15° flexion, before peaking at 3.25 ± 0.15 at 90° to exhibit a maximal 325% pressure increase, then dropping to 2.85 ± 0.14 at 120°. Pressure following lateral closing wedge osteotomy started at 1.14 ± 0.10 at 15° flexion, before peaking at 3.01 ± 0.13 at 90° to exhibit a maximal 301% pressure increase, then dropping to 3.00 ± 0.08 at 120°. Proximal opening wedge osteotomy cartilage pressure started at 1.54 ± 0.17 at 15° flexion, before peaking at 4.11 ± 0.16 at 60° to exhibit a maximal 411% pressure increase, then dropping to 3.16 ± 0.08 at 120°. The distal opening wedge osteotomy technique produced a cartilage pressure of 1.26 ± 0.13 at 15° flexion, before peaking at 3.35 ± 0.07 at 90° to exhibit a maximal 335% pressure increase, then dropping to 3.12 ± 0.12 at 120° (Fig. 4).

Fig. 4
figure 4

Relative patellofemoral pressures at 15°, 30°, 60° and 90° flexion following lateral closing HTO, proximal tuberosity medial opening HTO, and distal tuberosity medial opening HTO. Both distal tuberosity medial opening HTO and lateral closing HTO did not show a significant difference (P > 0.05) to the intact knee at all flexion angles. In contrast, the proximal tuberosity medial opening HTO technique showed a significant increase in patellofemoral pressure at 30° (P < 0.05), 60° (P < 0.005), and 90° (P < 0.0005) knee flexion. All pressures presented as mean pressure relative to the intact knee at 15°, designated a value of 1. All values presented as mean ± SEM. *P < 0.05, **P < 0.005, ***P < 0.0005

Full size image

The cartilage pressure following PTO opening wedge technique was significantly increased at 15° (P < 0.05), 30° (P < 0.005), and 60° (P < 0.0005) flexion when compared to the intact knee. The difference was not significant at 90° and 120°. In contrast to the proximal opening wedge technique, cartilage pressure following distal opening wedge osteotomy and closing wedge was not significantly (P > 0.05) different to the intact knee at all flexion angles.

Discussion

OWHTO is commonly accepted as a clinically effective procedure, but is plagued by postoperative complication. Specifically, patella infera following medial OWHTO with proximal tibial osteotomy (PTO) can predispose to anterior knee pain, crepitus, and reduced range of motion [12, 18, 23, 36, 53, 56, 64]. Although technical refinements [21, 24, 50, 56, 62, 64] have been trialed to decrease patella infera incidence, they are subject to considerations such as the size of correction and their effect on union [50]. Furthermore, PTO OWHTO produces a significant alteration in cartilage pressure, a factor linked to the degeneration of joint cartilage [2, 5, 10, 59]. This paper has elucidated that whilst PTO OWHTO produces a significant elevation in cartilage pressure, distal tibial osteotomy (DTO OWHTO) and closing wedge osteotomy produce no significant patellofemoral pressure change compared to the intact knee.

Many surgeons favor a lateral closing wedge HTO for treating varus tibiofemoral malignment, with up to 71% good results reported at 8 years follow-up [11, 31, 42]. But it should be noted that the procedure is limited by complications such as medial collateral ligament instability, patella infera, loss of correction, delayed union and nonunion, fracture, and arthrofibrosis [7, 16, 26, 33, 44, 48, 54, 62, 63]. Whilst complications following medial opening wedge HTO have also been illustrated, the frequency appears reduced and clinical outcomes are generally comparable to the closing wedge technique [1, 25, 26, 40, 41, 43, 60]. However, opening wedge osteotomy is clearly set apart by its reduced comparative invasiveness; with the closing wedge procedure relying on extensor muscle separation for the fibular osteotomy, and the risk of peroneal nerve palsy following fibula osteotomy. Asides from bypassing these issues, opening wedge osteotomy has the additional benefit of allowing coronal and sagittal corrections intraoperatively to optimize loading axis transmission through the knee. In order to clarify the more effective technique, the authors believe more research into comparative outcomes and complications such as patella infera are warranted.

Patella infera might not only accelerate cartilage degeneration through an increase in patellofemoral load, but also through a change in normal knee kinematics. As suggested by Andriacchi et al. [5, 10], this loss of load-bearing homeostasis in PTO OWHTO may translate to an inferior shift of patellofemoral pressure from an area of high loading and acclimatized coefficient of shear friction, to an area of low loading, causing mechanical injury to the newly loaded cartilage and inferring an exacerbated coefficient of shear friction that heightens the susceptibility to degeneration [5]. Given the fact that joint areas with low contact pressure and infrequent load-bearing are more sensitive to mechanical injury and cartilage loss [5, 8, 10], this inferior alteration in joint kinematics may have a two-fold effect on degeneration of the joint surface. It is likely that not only would the newly loaded area be susceptible to degeneration, but adhering to Wolff’s Law and the well known compression-induced maintenance of articular cartilage matrix metabolism, the previously loaded area may experience cartilage degeneration by aberrant subchondral bone remodeling and/or an imbalance between the anabolic and catabolic regulators of cartilage matrix. Indeed, various studies have documented the link between abnormal pressure and osteoarthritis [8, 10, 17, 32, 47, 58, 61].

Additionally, patella infera has been documented by several groups as a common complication of OWHTO for valgus correction of medial compartment osteoarthritis in the knee [18, 49, 53, 56, 63, 64]. Patella infera and early tendofemoral moment are thought to commonly result in anterior knee pain after conventional OWHTO. It should be noted however that whilst anterior knee pain is reported as an OWHTO complication within the literature, Mont et al. [46] found that patella infera did not affect clinical outcome, although the criteria used to determine patella infera (Insall-Salvati ratio less than 1.0) was less stringent than used by Windsor et al. [63] who reported 80% patella infera in their osteotomized cohort. Whilst the clinical significance of patella infera may not be well established [25, 53, 64], the exaggerated technical difficulty of total knee arthroplasty (TKA) subsequent to HTO certainly is. To this end, numerous authors have commented on the difficulty of revision to TKA after HTO, with concern mainly surrounding deformities of the proximal part of the tibia, patellar eversion, and soft tissue imbalances [4, 9, 27, 34, 37, 46, 57, 63]. Whilst further research is needed to confirm any conclusions drawn from this study, we believe distal opening wedge osteotomy or closing wedge osteotomy may reduce or prevent many of these complications, particularly deformation of the proximal tibia.

Whilst the findings of this study are clinically important to the safety and efficacy of correcting valgus tibiofemoral deformities, there are two limitations in this experiment worthy of discussion. Firstly, previous studies have predominantly used Tekscan pressure-sensitive films as a measurement device [2, 15], while this study used a Pedar™ system. The main difference in these two systems is that Tekscan films are 0.1 mm thick, whereas the Pedar™ sensor system is closer to 1 mm. The exact consequence of this increased sensor thickness is unclear, but we believe the raw pressure data in the present study has been altered to a small degree through the thickness of the sensor system. But in acknowledging this fact, our raw pressure data was similar to previously published data with Tekscan film [15, 28, 38, 39]. Secondly, the angle of correction made in this study was relatively large compared to that commonly observed [12, 13, 20, 53]. Gaasbeck [18] recently reported that the larger the correction angle made in OWHTO, the lower the resulting patella height. Therefore, we would expect a quantitatively smaller effect on patellofemoral cartilage pressure in smaller correction procedures, a hypothesis consistent with that of Mont et al. [45].

A successful HTO outcome relies on proper patient selection, stage of arthrosis, and the precision and maintenance of adequate operative correction [6, 25, 44, 48, 52]. Regarding the precision of the correction, we see an indication for lateral closing wedge osteotomy in patients with severe osteoarthritis of the medial compartment combined with an increased posterior tibial slope. A medial opening wedge osteotomy above the tuberosity should be considered only in cases with patella alta. In all other cases, we favor the medial opening wedge technique with a biplanar osteotomy below the tibial tuberosity (DTO). Additionally, larger correction angles are not contraindicated for the opening osteotomy, since an increase in medial compartment pressure can be avoided by releasing the superficial portion of the medial collateral ligament. Importantly, the deep portion of the medial collateral ligament (the stabilizing portion) remains intact. In the case of a concurrent anterior cruciate ligament reconstruction, the opening wedge osteotomy has the advantage of being quicker than the lateral closing wedge osteotomy, although one should take care not to increase posterior slope in this case since this would increases stress on the reconstructed ligament [14].

Given the relative advantages of DTO OWHTO compared to the traditional lateral closing wedge osteotomy (less invasiveness, shorter operation time, higher precision for correction, no risk of peroneal nerve palsy) [35, 40] and to the PTO OWHTO (maintenance of patellofemoral cartilage pressure with the possible prophylactic effect of this on cartilage degeneration), we recommend this technique for valgus correction of medial compartment osteoarthritis, specially in those with pre-existing anterior knee pain, either due to patella infera or patello femoral arthritic changes.