To our knowledge, this is the first study with a large sample size to develop a model for the prediction of risk of requiring TKA based on dynamic changes in two indirect radiological indicators: FH and misalignment of the lateral knee joint. In this study, we found that larger ΔFH and ΔLPW were associated with shorter remaining life of the knee joint and greater risk of requiring TKA. ΔMPTA and ΔJLCA were positively correlated with risk of TKA (r = −0.523, − 0.480, all P < 0.001), but these correlations were not statistically significant in the predictive model of the remaining life of the knee joint. The changes in FH and LPW indicate lateral knee joint degeneration, while changes in MPTA and JLCA indicate varus deformity and narrowing of the medial knee compartments.
Previous studies have shown that willingness to consider joint replacement, socioeconomic status, general health status, arthritis severity, educational level, age, and sex are important factors determining the utilization rate of TKA[9–11]. Willingness to consider joint replacement appears to be the most dominant factor associated with utilization of TKA; low socioeconomic status, poor general health status, low educational level, and female sex are associated with low TKA utilization rates. Lower limb disability and high TKA utilization rate are associated with substantial social and economic burdens. Therefore, delaying OA progression and reducing the need for TKA are important but challenging issues. Defining the maximum remaining life of the knee joint (i.e., the risk of TKA and residual life of the knee joint) will help to improve patient awareness and compliance, preventing the need for TKA via early medical intervention, which will help to reduce the social costs of severe knee OA.
Here, we proposed the use of two indicators of lateral knee degeneration for prediction of disease progression and remaining life of the knee joint; our model was confirmed to be effective and reliable. X-rays were used to calculate the rate of change in vertical distance between the lateral tibial plateau and the fibular head, as well as the rate of change in lateral dislocation of the tibial plateau. This procedure is economical, convenient, and time-saving; it effectively quantifies lateral knee compartment degeneration in varus knee OA.
Zhang reported that fibular cortical support is an important reason for uneven pressure distribution between the lateral and medial knee compartments, which leads to a much higher incidence of varus knee OA than valgus knee OA [17]. A reduced medial proximal tibial angle is commonly found in varus knee OA, suggesting that the medial tibial plateau is compressed and lower than the lateral tibial plateau. Although the lateral tibial plateau is supported by the fibula, it can also exhibit slow settling during the progression of OA. Therefore, a rapid decrease in FH is positively correlated with OA progression. Long-term intraarticular pressure causes knee joint degeneration [18]. We refer to the wear of the medial knee compartment as destruction caused by great pressure; and degeneration of the lateral knee compartment is caused by the accumulation of long-term pressure. Degeneration of the knee joint is accelerated when the load-bearing capacity of the knee is insufficient to withstand the applied pressure. In the early stages of varus knee OA, pressure in the medial knee compartment is unbalanced with tibial plateau load-bearing capacity but does not involve the lateral tibial plateau; in the advanced stage of varus knee OA, imbalance in the lateral knee compartment is also present. Because the remaining life of the knee joint is related to the rates of change in the two lateral degeneration indicators, but not the rates of change in the medial indicators, the risk of requiring TKA is related to the relationship between lateral intraarticular pressure and load-bearing capacity.
We presume that the changes in lateral knee morphological characteristics observed in this study contribute to knee OA progression by aggravating the knee joint instability. Theoretically, the function of the posterolateral ligament complex (i.e., maintaining the normal force line of the knee and the normal tibial rotation angle) declines during upward shift, because it relaxes. First, the pressure in the medial knee compartment increases when knee varus deformity occurs [19, 20], which eventually leads to knee degeneration (e.g., hyperosteogeny and osteosclerosis). Simultaneously, an abnormally increased tibial rotation angle is associated with increased pressure in the lateral knee compartment [19, 20]. Therefore, in addition to reflecting bone degeneration, the fibula shifts upward relative to the tibial plateau; this manifests as a decrease in FH, which could also indicate accelerated functional degeneration of muscles and ligaments. The relative outward displacement of the tibial plateau, observed as an increase in LPW, represents increased lateral structural stress; this reflects the lack of medial and lateral constraint strength, which affects the state of the fibula head [21, 22]. An increase in LPW may be the direct cause of increased stress in lateral structures when varus knee OA occurs.
To our knowledge, there is no clear consensus regarding indications or appropriateness criteria for TKA [10]. Generally, when OA only occurs in the medial compartment, unicondylar knee arthroplasty or high tibial osteotomy can be considered [23–26]. When both medial and lateral compartments are involved, TKA is chosen. Therefore, the degeneration of lateral knee structures appears to be more closely related to the use of TKA for varus knee OA. Accelerated deterioration of the disease is often accompanied by significant symptomatic or functional deterioration, which could increase a patient’s willingness to seek medical treatment or undergo TKA [11]. Therefore, the rates of change in FH and LPW can help surgeons to distinguish between mild and advanced knee OA; this helps to reduce the incidences of missed diagnosis and misdiagnosis. Patients with high risk of requiring TKA can be differentiated from patients with mild knee OA alone; early medical intervention (e.g., weight loss, personalized exercise, and knee-preserving surgery) may enable postponement or prevention of TKA, thus reducing the burdens on healthcare and medical insurance systems.
The data presented here are clinically advanced in the following four respects. First, X-ray examination is economical and convenient, greatly improving the rates of patient examination and re-examination; this facilitates knee OA evaluation based on dynamic changes. Second, all patients enrolled in this study had TKA indications, such as K-L grade IV, severe pain, severe limitation of activity, and poor quality of life. However, they had been treated with TKA, thus avoiding some bias (e.g., low rate of TKA utilization because of low socioeconomic status, poor general health status, and/or low educational level). Third, risk of requiring TKA can be shown intuitively by this mathematical model which helps us to identify patients with rapidly aggravated knee OA and intuitive and pellucid result is helpful to improve patients' compliance by improving their understanding of knee OA and TKA. Early medical intervention will be carried out strictly for these patients to delay the progress of OA. Fourth, for varus knee OA (considering medial knee joint degeneration alone), unicondylar knee arthroplasty, high tibial osteotomy, and TKA could all be chosen; however, these lack clear appropriateness criteria. The use of indicators of lateral knee joint degeneration could facilitate more rational use of TKA.
This study had several limitations. First, patients without records of TKA after the longest follow-up period in this study were excluded; this ensured that the study focused only on TKA risk, rather than the threshold for determining whether TKA surgery will be performed in the future. Therefore, additional systematic studies are required. Second, retrospective research can explore correlations in data and build a correlation model based on random sampling. However, to make the predictive model more convincing, additional prospective studies are needed to verify its sensitivity and reliability by comparing differences between theoretical and actual values. Third, although the total amount of data was large, a small amount of data met our study criteria, reflecting incomplete follow-up because of patient concerns regarding X-rays or a lack of appropriate understanding concerning the seriousness of knee OA.