In radiographic studies of knee OA, the lateral anatomical and morphological characteristics of the knee have not received sufficient attention, possibly because lateral degeneration seems to be less severe than medial degeneration or because the fibula is not the main component of the knee joint and bears little axial load [13, 14]. However, lateral structures play a role in degeneration of the knee; the bony support of the fibula has a role in the mechanism of knee OA, which was named the “nonuniform settlement” theory [8, 9]. The soft tissues of the lateral knee are important for knee stability. The posterolateral ligament complex, including the lateral collateral, arch, and popliteal ligaments and popliteal tendon, prevents varus deformity and abnormal external rotation of the tibia [15–20]. These morphological abnormalities are obvious in knee OA [21–23]. Therefore, lateral knee structures deserve more attention, especially in knees with OA.
To our knowledge, this is the first large study to describe the anatomy and morphology of the lateral knee joint and its association with knee OA. Our results showed that in varus knee OA, the anatomical position and morphology of the lateral knee on radiography changed, and may be determined by changes in the mechanical environment of the lower limbs. These changes were accompanied by changes in K-L grade and the morphological indexes MPTA, JLCA, and HKAA, suggesting that LPW and PFC are significantly and positively correlated with the severity of knee OA and genu varus deformity, while FH is negatively correlated.
LPW is sensitive to knee alignment and indicates the relative location and distance between the proximal lateral tibia and lateral femoral condyle by drawing two lines perpendicular to the medial tibial articular surface . Studies have used LPW to evaluate lateral tibial plateau fractures . We found that LPW can also be used to evaluate the degree of knee alignment in patients with varus knee OA. The average LPW of a healthy knee is 0.02 ± 2.03 mm, so the lateral aspect of the tibial plateau should be collinear with the lateral femoral condyle [24, 25]. Here, almost all LPW measurements in patients with varus knee OA were positive. The proximal lateral tibia was more lateral than the femur and was closely related to knee OA occurrence, progression, and varus deformity. Because of the relative outward displacement of the tibial plateau, a positive LPW represents an increase in stress in the lateral structure, which in turn affects the state of the fibula head [24, 25]. A change in LPW might be the direct cause of increased stress in the lateral structures when varus OA occurs.
The PFC describes morphological variation in the fibula instead of the fibular shaft axis. Kuroda et al. defined the fibular shaft axis as the line connecting the center of the fibular head to the center of the lateral malleolus; it is commonly used as a radiographic landmark for planning total knee arthroplasty . PFC was defined as the angle formed by the proximal medullary cavity central line and the middle medullary cavity central line of the fibula in a previous study with inaccuracy because the curvature was described by angle. PFC reflects the degree of bending in the proximal fibula with a new definition used here and calculated by the formula mentioned above, and it is proportional to lateral stress. We found that the PFC was significantly correlated with JLCA, MPTA, and HKAA and directly or indirectly affected the lateral structure of the knee, similar to the fibular shaft axis on plain radiographs. Xie et al. reported that the medial cortex of the proximal fibular shaft is a reliable landmark for the mechanical axis of the tibia in positive HKAA knee OA . Our results suggest that the validity of the fibular shaft axis may be an objective manifestation of changes in mechanical stress on the fibula, which are influenced by changes in the load along the tibial mechanical axis. This suggests that changes in the mechanical axis of lower limbs contribute to changes in both the tibia and fibula, rather than just one in isolation. Because of the proximal tibiofibular joint structure, little dislocation of the fibula occurs in the presence of genu varus deformity. The lateral muscles and tendons place more tractive force on the lateral fibular cortex, as demonstrated by the periosteal reaction in the lateral cortex of the fibula on OA radiographs. This mechanism may be the main cause of fibula bending.
We proposed to use the parameter FH to describe the height of the fibular head relative to the tibial plateau. The decrease in FH observed in knee OA indicates that the fibula shifts upward. FH was significantly and negatively correlated with K-L grade (odds ratio [OR] = 0.53), JLCA, and HKAA, and was significantly and positively correlated with MPTA. According to Preuschoff, the vertical force acting on the fibula is a tensile force, not a compressive force. This explains the upward movement of the fibula, although it may also result from settling of the lateral tibial plateau, which should be examined in further studies. LaPrade et al. reported that the fibula head affects the force and proximal tibiofibular articulation via the lateral structures, which transmit and distribute the lateral tensile force generated by limb gravity . Theoretically, the function of the posterolateral ligament complex declines as the fibula shifts upward because it relaxes. Simultaneously, the pressure in the medial knee compartment increases [29, 30], which eventually leads to knee degeneration, such as hyperosteogeny and osteosclerosis . Consequently, the decreased FH could be a lateral knee characteristic that aggravates knee OA.
The interaction between the fibula and tibia may be involved in adaptive changes in the pathogenesis and progression of knee OA, but this is currently unclear. Lacking superior or inferior bony support, the fibula cannot bear gravity directly but anchors to the tibia via the interosseous membrane and proximal and distal tibiofibular joints. Considering the structural proximity of the tibia and fibula, factors influencing fibula changes in OA might include increased traction of the lateral knee joint and proximal tibia varus deformation on the fibula, which would gradually increase in prominence with knee OA severity. There is extensive evidence to show that changes in the tibia's anatomical and mechanical features affect the fibula's position and morphology, especially the fibula position [32–34]. The traditional view is that the main function of the fibula is to dissipate and transmit axial loads during weight-bearing, but recent studies have suggested mechanical self-adaptability of the fibula in OA and the influence of the fibula on the tibial load in turn [35–37]. Therefore, changes in the anatomical and morphological characteristics of the fibula and tibia in OA may help to explain the mechanical interaction between the tibia and fibula during knee OA.
Generally, varus knee OA involves medial knee compartments first and gradually involves the lateral. When medial and lateral knee compartments are all involved which is the end-stage osteoarthritis, total knee arthroplasty is the major surgical treatment. Therefore, it is not enough for traditional indicators to focus only on the degeneration of the medial structures and the lack of a method for evaluating degeneration of the lateral knee structures may be an important reason for the lack of standard indications for total knee arthroplasty. Consequently, parameters describing the changes in lateral knee structures with OA will enable a more comprehensive assessment of varus knee OA severity and deformity. Furthermore, it will provide us ideas for predicting the risk of total knee arthroplasty.
This study has several limitations. First, this study involved a large-sample radiographic analysis, so biomechanical findings were neglected. Imaging variation in the fibula does not directly reflect mechanical changes around the knee joint. Second, the rotational relationship between the tibia and fibula cannot be measured using plain radiography of the knee, although this relationship may exist in knee OA. Third, the impact of rotation or flexion of the limbs on the measurements cannot be observed on 2D radiographs.