Consistent with the findings of Gross et al[7–11], our study confirmed the clinical phenomenon that the incidence of osteoarthritis or articular cartilage damage of the medial patellofemoral compartment(123cases, 50.0%) was higher than that in the lateral compartment (70 cases, 28.5%)[5, 12–14]. There are currently two explanations for the high incidence of the medial patellofemoral osteoarthritis: disuse degeneration of the surface cartilage of the patellofemoral joint[9] and the reaction force of the lateral patellofemoral surface leading to medial patellofemoral wear[10]. Stefanik and Gross agreed that the incidence of cartilage damage in the medial patellofemoral was higher than in the lateral but did not support the notion that the high incidence of medial patellofemoral cartilage damage was due to disuse degeneration. There should be no subchondral bone marrow lesions in disuse degeneration of articular cartilage, which is usually due to no stress. They found that subchondral bone marrow lesions in the medial patellofemoral compartment are also relatively common in their multicenter clinical studies[7, 11].Gorniak similarly observed more cartilage damage in the medial patellofemoral compartment than in the lateral side by cadaveric dissections of the patellofemoral joint, which he interpreted to mean that both the medial and lateral sides of the patellofemoral joint have their specific patterns of articular wear and that the wear of the articular surfaces of the medial patellofemoral compartment may be due to the increase in the reaction forces of the medial patellofemoral joint resulting from the lateral patellofemoral compartment taking up the stresses[10]. Nevertheless, the literature suggests that the contact area and pressure on the articular cartilage of the lateral patellofemoral compartment is greater than that of the medial compartment; the incidence of articular cartilage damage in the lateral compartment should be more than in the medial; in contrast, the incidence of articular cartilage damage in the medial compartment of the patellofemoral was significantly higher than that in the lateral[7, 11, 16, 38, 39]. After mapping the pattern of patellofemoral cartilage damage by collecting MRIs at 3.0 T in people aged 30 to 79 years without knee symptoms, Katano et al. concluded that medial patellofemoral cartilage damage was more than that of the lateral side and that the pattern of cartilage damage was in the direction of the trochanteric crest of the femur[8]. This study indirectly suggests that damage to the medial patellofemoral cartilage is caused by mechanical stressors and not disuse atrophy. In conclusion, the possible mechanism for the high prevalence of osteoarthritis in the medial patellofemoral compartment is even more puzzling compared to the mechanism for developing osteoarthritis in the lateral patellofemoral compartment.
Many clinical studies and epidemiologic investigations have concluded that the incidence of osteoarthritis in the lateral compartment of the patellofemoral joint is higher than in the medial one, which has been attributed to the fact that their investigations have been performed by X-ray rather than MRI[2, 23, 40]. Studies have shown that MRI can accurately identify early joint degeneration and cartilage damage [26–28], and Hunter's proposed MRI diagnostic criteria for osteoarthritis based on MRI [12]. The WORMS score can be compared with good agreement between osteoarthritis of the medial and lateral compartments of the patellofemoral joint[13].In the current studies of cartilage damages or patellofemoral stress to the medial and lateral compartments of the patellofemoral joint, the effects of varus and vaglus alignment and rotation of the lower extremity on patellofemoral joint stresses are more commonly studied.
Cahue demonstrated through an 18-month longitudinal follow-up study that varus alignment of the lower extremity (with a neutral hip-knee-ankle angle of 0°, divided into varus and valgus) increased the odds of progression of isolated medial patellofemoral osteoarthritis, whereas valgus alignment of lower extremity increased the odds of progression of isolated lateral patellofemoral compartment osteoarthritis but with a weaker correlation[40]. The Elahi study showed that the presence of knee varus was more likely to be present in combination in patients with osteoarthritis of the medial compartment of the patellofemoral joint[23]. The above two studies seem to affirm that knee inversion increases stress on the medial patellofemoral compartment and thus leads to osteoarthritis of the medial patellofemoral compartment. However, the following studies give the exact opposite conclusion. Otsuki concluded after observation that most of the patellofemoral arthritis in patients with varus lower limb was concentrated in the lateral compartment of the patellofemoral joint[25]. Shuntaro's study concluded that an inversion deformity of the knee may lead to an outward displacement of the tibial tuberosity and an outward tilt of the patella, which in turn may lead to osteoarthritis of the lateral patellofemoral intercompartment[24]. The Iijima study noted a significant correlation between varus thrust from knee varus and a higher probability of osteoarthritis in the medial compartment of the patellofemoral joint[41].The Gross study, on the other hand, demonstrated that medial compartment cartilage damage in the patellofemoral joint was highly prevalent in all lower extremity line of force categories, whereas only in more severe knee osteoarthritis was the articular cartilage damage of the lateral compartment patellofemoral of the valgus knee greater than or equal to medial compartment osteoarthritis[7].The results of the Hinterwimmer study also concluded that no patellofemoral kinematic alterations in patellofemoral motion and patellofemoral joint contact area were found in patients with knee entropion and mild medial osteoarthritis[42].In line with the findings of Hinterwimmer, our findings similarly showed no correlation between lower extremity alignment, either HKA varus or valgus, and the incidence of osteoarthritis in the medial and lateral compartments of the patellofemoral joint.( Table 4 and Table 6)
Although it has been reported in the literature that knee varus can lead to an outward displacement of the tibial tuberosity[24]. It is taken for granted that knee varus can lead to an inward displacement of the tibial tuberosity and a decrease in the Q-angle, and that a decrease in the Q-angle decreases lateral patellofemoral articular surface pressures while increasing medial patellofemoral articular surface pressures. The results of Mizuno's study indicated that an increase in the Q angle could lead to patellofemoral Lateral dislocation or increased lateral patellofemoral contact pressures, whereas a decrease in Q angle may not displace the patella medially[43]. Stephen's study showed that gradual lateralization of the tibial tuberosity increased lateral contact pressures, increased lateral patellofemoral trajectory, and decreased patellofemoral stability, whereas medial contact pressures and tracking were also altered with gradual medialization of the tibial tuberosity, albeit with slight variations[44]. Mizuno's study showed that an increase in Q angle might lead to a lateral patellofemoral displacement or an increase in patellofemoral lateral contact pressures. Whether medialization or lateralization of the tibial tuberosity is produced by the varus and valgus of the lower limb force line remains controversial. Hence, the effect of the lower limb alignment on osteoarthritis of the medial and lateral compartments of the patellofemoral joint is still theoretically uncertain. Our study shows that neither varus nor valgus of the knee joint affects the incidence of medial or lateral patellofemoral osteoarthritis. Worlicek showed through the results of cadaveric patellar three-dimensional motion experiments that knee varus leads to patellar external rotation and patellar tilt angle is significantly greater than neutral and external and concluded that analysis of the tibiofemoral alignment alone is not sufficient to predict patellofemoral kinematics[45]. This may suggests that changes in the alignment of the lower extremity will also be linked to other factors that contribute to changes in pressure in the medial and lateral patellofemoral compartments and that the result is not necessarily an increase in pressure in the medial or lateral patellofemoral joint compartments.
Lee concluded that femoral and tibial rotation can significantly affect the pressures and contact area of the medial and lateral compartments of the patellofemoral joint, respectively[19]. After studying tibial relative to femoral external rotation, Miao concluded that tibial relative to femoral external rotation increases the risk of recurrent patella dislocations in children[26]. Takai concluded that Lower extremity rotational alignment is an important causative factor for knee osteoarthritis, including patellofemoral arthritis[46]. Kaiser showed that the patellofemoral joint's static patellar tilt and axial contact during knee extension were mainly affected by knee torsion, TT-TG, and trochlear dysplasia but not by femoral or tibial torsion[47]. In this study, multifactorial logistic regression analysis demonstrated a correlation between femoral tibial rotation angle and osteoarthritis of the patellofemoral compartment, with more medial patellofemoral osteoarthritis occurring in the patellofemoral joint when the FTR > 10° (OR = 3.497, 95%CI:1.370 ~ 8.926, P = 0.009). When the FTR < 0° (OR = 4.601, 95%CI: 1.416 ~ 14.946, P = 0.011), there was more incidence of lateral patellofemoral compartment osteoarthritis. Both previous literature and the present study suggest that the femoral tibial rotation angle significantly affects patellofemoral compartment pressures.
Liao concluded that patella shapes in patellofemoral joint degeneration were associated with a hooked lateral patellar projection, evenly distributed facets, roundedness and thickness, and a larger patella[48]. The findings of Liao's study are similar to the present study, demonstrating that patellar shape correlates with the severity of patellofemoral osteoarthritis. Eijkenboom's study found that patellar shape was associated with the prevalence of PFOA in women based on magnetic resonance imaging and that patellar shape may be a risk factor for early detection of PFOA[49]. Dai's study showed that Wiberg type III patella were associated with higher severity of lateral patellofemoral osteoarthritis[22]. Our study concluded that Wiberg type I patella (OR = 4.364, 95%CI:1.910 ~ 9.969, P = 0.000) was a high-risk factor for the development of osteoarthritis of the medial patellofemoral compartment. Wiberg type III patella (OR = 2.916, 95%CI: 1.267 ~ 6.712, P = 0.012) were high-risk factors for the development of lateral patellofemoral compartment. The results of the present study have similarities with Dai[22], which may show that patellar profile can affect medial and lateral patellofemoral compartment pressures.
The most consistent changes in the knee joint with age are damage to articular cartilage and increased prevalence of joint disease[50, 51]. In the findings of this study, increasing age was an independent risk factor for osteoarthritis in the medial compartment of the patellofemoral joint but not about the lateral compartment of the patellofemoral joint. Force distribution on the patellofemoral articular surface during knee flexion involves a complex and dynamic interaction between quadriceps forces, soft tissue constraints, and patellar tracking in the bony geometry of the patellofemoral joint. Osteoarthritis of both the patellofemoral compartment joints worsens over time. The severity of osteoarthritis of the medial patellofemoral compartment may be more strongly correlated with change over time. In contrast, the severity of osteoarthritis of the lateral interphalangeal compartment is less strongly associated with change over time. In addition, univariate analysis showed increased patellar tilt angle, increased patellofemoral index, increased TT-TG, and lower K-L classification were independent protective factors for the development of osteoarthritis in the medial compartment of the patellofemoral joint. While females, high BMI were independent risk factors for the development of osteoarthritis of the lateral patellofemoral compartment, and congruence angle became larger was a independent protective factor for lateral patellofemoral osteoarthritis. All of the above factors were correlated with the development of osteoarthritis in the medial-lateral patellofemoral compartment, but not significant in multifactor regression.
Hirschmann found that the FTR angle changed from external to internal rotation, the TT-TG distance became smaller, and the patellar tilt angle became smaller in the upright weight-bearing position than in the supine position, indicating that the tibia rotated internally, the patella tilted inward relative to the femur in the upright weight-bearing position, and this difference was statistically significant[52]. The study by Hirschmann illustrates that the mechanical environment of the patellofemoral joint is quite different in the weight-bearing and non-weight-bearing positions. The contact stresses, and contact area of the medial compartment of the patellofemoral joint in the weight-bearing position is likely to be increased. The angle of lateral patellar tilt decreases with increasing tibial rotation relative to femoral internal rotation, and the stresses and contact areas on the medial surface of the patellofemoral joint increase[53]. Powers came to the same conclusions as the findings described above after examining changes in the alignment position of the patellofemoral joint in the weight-bearing situation[54]. This variation of knee parameters in natural environments reminds us to pay attention to the study of actual biomechanical conditions in weight-bearing and dynamic environments when studying the patellofemoral joint. Von Eisenhart-Rothe investigated a three-dimensional kinematic technique to find the samilar results, and they discovered that patellar tilt was reduced and the femur had a significant posterior displacement and external rotation relative to the tibia during knee flexion [55]. When analyzing the relationship between knee mechanics and patellofemoral osteoarthritis, one should not only rely on the static mechanics of the patellofemoral (ligaments and bones) but also include dynamic mechanics (musculo-neurological system) to analyze the changes in pressure in the medial and lateral compartments of the patellofemoral.
There are several limitations of this study: (i) Due to the retrospective study, hip-knee-ankle CT was missing in the data set, and the effect of femoral anteversion angle and Q angle on the change of patellofemoral joint surface pressure was not analyzed. (ii) The femoral, tibial rotation angle and TT-TG were measured in MRI in this study, and the results obtained may differ from those measured by CT. (iii) This study is a single-center study. The results of this study would be more convincing if the sample size could be increased and a multicenter study could be conducted.
Conclusion
FTR angle and patellar Wiberg type have a strong influence on the severity and incidence of medial and lateral osteoarthritis of the patellofemoral joint. The prevalence of MPFOA may not be related to varus and vaglus of the lower limb. The dynamic factors of the weight-bearing position of the patellofemoral joint are important when studying the pathogenesis of medial patellofemoral osteoarthritis.