Abstract
Introduction
High tibial osteotomy (HTO) is a commonly used surgical technique for treating moderate osteoarthritis (OA) of the medial compartment of the knee by shifting the center of force towards the lateral compartment. The amount of alignment correction to be performed is usually calculated prior to surgery and it's based on the patient's lower limb alignment using long-leg radiographs. While the procedure is generally effective at relieving symptoms, an accurate estimation of change in intraarticular contact pressures and contact surface area has not been developed. Using electromyography (EMG), Meyer et al. attempted to predict intraarticular contact pressures during gait patterns in a patient who had received a cruciate retaining force-measuring tibial prosthesis. Lundberg et al. used data from the Third Grand Challenge Competition to improve contact force predictions in total knee replacement. Mina et al. performed high tibial osteotomy on eight human cadaveric knees with osteochondral defects in the medial compartment. They determined that complete unloading of the medial compartment occurred at between 6° and 10° of valgus, and that contact pressure was similarly distributed between the medial and lateral compartments at alignments of 0° to 4° of valgus. In the current study, we hypothesised that it would be possible to predict the change in intra-articular pressures based on extra-articular data acquisition.
Methods
Seven cadavers underwent an HTO procedure with sequential 5º valgus realignment of the leg up to 15º of correction. A previously developed stainless-steel device with integrated load cell was used to axially load the leg. Pressure-sensitive sensors were used to measure intra-articular contact pressures. Intraoperative changes in alignment were monitored in real time using computer navigation. An axial loading force was applied to the leg in the caudal-craneal direction and gradually ramped up from 0 to 550 N. Intra-articular contact pressure (kg) and contact area (mm2) data were collected. Generalised linear models were constructed to estimate the change in contact pressure based on extra-articular force and alignment data.
Results
The application of an axial load results in axial angle changes and load distribution changes inside the knee joint. Preliminary analysis has shown that it is possible to predict lateral and medial compartment pressures using externally acquired data. For lateral compartment pressure estimation, the following equation had an R of 0.86: Lateral compartment pressure = −1.26*axial_force + 37.08*horizontal_force − 2.40*vertical_force − 271.66*axial_torque − 32.64*horizontal_torque + 18.98*vertical_torque − 24.97*varusvalgus_angle_change + 86.68*anterecurvature_angle_change − 17.33*axial_angle_change − 26.14. For medial compartment pressure estimation, the following equation had an R2 of 0.86: Medial compartment pressure = −2.95*axial_force −22.93*horizontal_force − 9.48*vertical_force − 34.53*axial_torque + 6.18*horizontal_torque − 127.00*vertical_torque − 110.10*varusvalgus_angle_change − 15.10*anterecurvature_angle_change + 55.00*axial_angle_change + 193.91
Discussion
The most important finding of this study was that intra-articular pressure changes in the knee could be accurately estimated given a set of extra-articular parameters. The results from this study could be helpful in developing more accurate lower limb realignment procedures. This work complements and expands on previous research by other groups aimed at predicting intra-articular pressures and identifying optimal alignment for unloading arthritic defects. A possible clinical application of these findings may involve the application of a predetermined axial force to the leg intra-operatively. Given the estimated output from the predictive equation, one could then perform the opening wedge until the desired estimated intra-articular pressure is achieved. With this method, an arthrotomy and placement of intra-articular pressure sensors would not be needed. This work is not without its limitations. This experiment was performed on cadaveric specimens. Therefore, we cannot directly predict what the pressures would be in a de-ambulating patient. However, these sort of experiments do help us understand the complex biomechanics of the knee in response to alterations in multi-planar alignment. Further in vivo research would be warranted to validate these results. Additionally, given our current experimental setup, only axial loading could be performed for testing. Further experiments involving dynamic motion of the lower limb under load would further help us understand the changes in pressure at difference flexion angles. Continued experiments would help us gather additional data to better understand the relationship between these variables and to construct a more accurate predictive model. In summary, we have established a framework for estimating the change in intra-articular contact pressures based on extra-articular, computer-navigated measurements. Quantifying the resulting changes in load distribution, alignment changes, torque generation and deflection will be essential for generating appropriate algorithms able to estimate joint alignment changes based on applied loads.
