Patellofemoral joint geometry and dyskinesia

Alterations in patella height (alta and baja), tilt and displacement have been associated with joint space narrowing, cartilage loss and bone marrow lesions ( Fig. 4 ). At the femoral side of the articulation, a shallow trochlea, particularly the lateral trochlea surface, has also been associated with increased odds PF cartilage damage and lateral PF compartment bone marrow lesions . The relationships between the type of alteration in PF joint geometry and presence of PF OA differs markedly between the medial and lateral PF compartments. Lateral displacement of the patella, shallow trochlear groove and decreasing lateral patella tilt have all been linearly associated with the magnitude of joint space narrowing, cartilage damage and bone marrow lesions in the lateral compartment . In one study, the odds ratio between lateral patella displacement and lateral joint space narrowing and lateral PF osteophytes indicated an 8- and 3-fold increase in the odds respectively ( Fig. 3 a). In contrast, increasing lateral patella tilt is the only pathomechanical characteristic that has been consistently linearly related with medial joint space narrowing and cartilage damage (there were no correlates with medial bone marrow lesions) ( Fig. 3 b). This has important implications for treatment design, as even small correction in PF joint geometry could reduce the odds of lateral PF OA damage, however this is not the case for those with medial compartment disease.

Schematic demonstrating measurements of patellofemoral malalignment indices. (a) Patella height measured by Insall-Salvati ratio = length of patella tendon (b): superior–inferior length of the patella (a), with examples of patella alta and baja. (b) Medial–lateral patella displacement relative to the femur as measured by the bisect offset (c) Trochlear groove depth as measured by the sulcus angle, which is defined by joining the highest points of the medial and lateral femoral condyles with the lowest point of the intercondylar sulcus or, when the groove lacks discernable depth, defined by a perpendicular line projected anteriorly from the posterior condylar line. (c) Lateral patella tilt measured as the angle between the patella and posterior femoral condyles.

It is theorised that increased mobility of the patella, by being altered in shape and not confined to the trochlear groove, increases shear forces and joint stress . Farrokhi et al. provided evidence that peak and mean joint pressure and shear stress at the patella and femur are increased when individuals with PFPS performed a squatting task. The authors theorised that shear stress reflects the portion that tends to distort joint tissue and is the possible risk factor for cartilage breakdown. There is little in vivo evidence for the specific patella and femoral geometry risk factors identified above. However, in cadaveric knees, experimentally induced lateral displacement and tilt of the patella significantly increased mean lateral facet pressure . This effect was strongest near extension. A similar effect was reported in a mechanical model study of the biomechanical effect of patella height. Patella baja significantly increased patella contact pressures in knee extension whereas patella alta increased the magnitude and duration of patella contact forces during flexion . While the latter study did not specify whether the medial or lateral compartment was experiencing greater stress, the studies suggest that individuals displaying altered PF geometry characteristics may be experiencing increased joint stress at times when the PF joint is not normally loaded.

Muscle force and activation

The quadriceps, hip abductors, gluteals, hamstrings and iliotibial band (ITB) have all been implicated in the presence of increased PF joint stress. The majority of research has focused on quadriceps strength, as the peak PF joint reaction force is primarily influenced by the knee flexion angle and magnitude of quadriceps force . Hart et al. reported cross sectional areas of the vastii and rectus femoris were reduced in individuals with PF OA, implying that their force generating capacity was reduced. This findings has been supported by numerous studies indicating that individuals with PF OA negotiate stairs with decreased quadriceps force , that quadriceps weakness was positively associated with lateral cartilage damage and bone marrow lesions and that strong quadriceps reduce the odds of the presence of lateral PF OA ( Fig. 3 a). Interestingly, these associations were not found for medial PF OA. The controversy in these findings, however, is their relationship to joint stress and OA symptoms. Evidence regarding the association between peak quadriceps force and pain is inconsistent, with one paper finding no relationship and another concluding there is an interconnected relationship .

Of the remaining muscles, only hip abductors and gluteal force have been studied in individuals with PF OA . In these studies it was found that individuals with PF OA, produce significantly less gluteus medius and minimus force than healthy controls during level walking and descending stairs . They also exhibit significantly reduced hip abductor strength . The potential increase in femoral internal rotation could result from decreased hip abductor strength may lead to an increase in lateral displacement of the patella in the trochlear groove. In cadaveric models, increases in lateral displacement, produced by increased load on hamstrings and ITB, have been reported to shift force and pressure toward the lateral patella facet and increase joint stress by a minimum of 5% .

Patella flexion also increases with tight hamstrings as the tibia and patella tendon are translated posteriorly, subsequently increasing PF compression forces – even more so when combined with weakened quadriceps . Similarly, a tight ITB can pull on its lateral retinaculum attachment and increase lateral tilt of the patella . In a study of 16 healthy men, those with tight hamstrings exhibited significantly greater lateral PF compartment joint stress and significantly reduced medial PF compartment contact area during a squat task . While these studies offer an explanation for how specific muscles contribute to PF joint stress, results are not in vivo or in knees with PF OA damage. Forces applied by muscles in vivo, particularly the ITB, are unknown. Therefore, cadaver studies may over or underestimate the amount of force required in their experiments and may not have clinically meaningful consequences. Ward and Powers examined contributions to PF joint stress in a PFPS population and found results that differed considerably to previous cadaveric studies. Further, the magnitude of PF joint stress may differ between healthy individuals and those with PF joint pain in some tasks, such as walking , but not others such as negotiating stairs . This suggests that data from modelling studies and healthy cohorts may over or under represent the implications for PF OA populations and should be interpreted with caution.

Lower limb alignment and kinematics

Varus or valgus alignment, tibiofemoral rotation and foot morphology have all been cited as contributors to PF OA due to their potential influence on the Q-angle. The Q angle adds a strong, laterally directed component to the PF contact force, resulting in the majority of medial-lateral PF joint load to be on the lateral compartment . Several authors have reported that valgus alignment was associated with between two- and four-fold increase in odds of lateral PF OA and varus alignment increased the odds of medial PF OA ( Fig. 3 a, b). A simple mechanical explanation is that valgus decreases the Q-angle, increasing the lateral PF forces and varus alignment increases the Q-angle and thus increases medial PF forces. McWalter et al. added to this explanation by reporting patella kinematics (patella flexion, anterior translation and spin) differ during knee flexion depending on whether an individuals with knee OA exhibit varus or valgus alignment. However, these authors also observed inconsistencies in patella tilt and lateral translation within malalignment groups, suggesting that correction of malalignment may not necessarily correct the force distribution at the PF joint. Several other cross-sectional studies report medial PF bone marrow lesions and cartilage damage is prevalent in individuals with valgus and neutral alignment, as well as those in varus . Thus, the summation of evidence suggests that other alignment and kinematic factors need to be considered.

The influence of tibia and femur alignment and kinematics on PF mechanics has been well documented in cadaver models and individuals who are healthy or have PFPS. Individuals with PFPS have been found to exhibit significantly increased femoral internal rotation in knee extension and flexion compared with healthy controls . Increases in patella lateral tilt and translation were also observed in this study and in a study of healthy individuals who landed from a jump in increased femoral internal rotation . However, the magnitudes of internal femoral rotation reported in these studies were shy of the reported 20°–30° threshold required to increase joint contact pressure in cadaver models . Thus, it appears the hypothesis of Powers et al. that the primary contributor to lateral patella instability is femoral rotation, may only be valid if there is co-existing tibial external rotation.

External tibial rotation increases lateral patella shift and tilt in cadaver models and increases PF contact pressure in the lateral compartment . Combined tibiofemoral rotation independently explains 29% of variance in PF joint contact area in individuals with PFPS . Similarly, increasing internal tibial rotation increased contact pressure on the medial PF compartment, although increases are much less than those experienced in the lateral compartment . The lack of large influence of tibial internal rotation on PF compartment pressure may explain why the presence of pes planus was not associated with medial or lateral PF cartilage damage , despite the coupling of rearfoot eversion and internal tibial rotation. The lack of association may also be due to data being collected in static positions given that rearfoot-tibia coupling is most evident during running gait .

Patellofemoral joint geometry and dyskinesia

Similar to cross-sectional data, a longitudinal study demonstrated lateral displacement of the patella was linearly associated with odds of increasing lateral PF joint space narrowing over 35-month period ( Fig. 5 ). Patella alta also linearly increased the odds of progression of medial and lateral cartilage damage and lateral bone marrow lesions even when the influence of lateral patella tilt and displacement were adjusted for . In contrast to cross-sectional data, where large lateral patella tilt increased the odds of exhibited medial PF joint space narrowing, the same relationship was not found for studies of PF OA progression. A single study reported that increasing lateral tilt had a linear protective effect on medial PF joint space narrowing and no effect on narrowing in the lateral compartment. There are several theories that could explain this contrast. Increasing patella tilt could reflect laxity in the structures that hold the patella in a lateral tilt resulting in greater mobility and shear stress on the medial compartment. However, as PF OA progresses, it is unlikely that a single source of stress would be present in isolation. Thus concomitant tightening of the ITB or altered patella kinematics may influence patella tilt in a manner that reduces stress on the medial compartment.

Odds ratios and 95% confidence intervals for the risk of PF OA progression for specific pathomechanical characteristics based on longitudinal data. An odds ratio of 1 represents equal odds of progression. BML = bone marrow lesion, CD = cartilage damage, CL = cartilage loss, JSN = Joint space narrowing, Osteo = osteophytes.

A further difference between medial and lateral PF OA progression was their relationship with the depth of the trochlear groove. When individuals exhibited progression in lateral PF joint space narrowing, their sulcus angle was more likely to be in the first and fourth quartile, indicating the deepest or shallowest trochlear grooves . Whereas the greatest odds of medial PF OA progression were for those whose trochlear groove depth placed them in the middle quartiles. While there was no correlation between trochlea groove depth or lateral patella displacement and pain , this indicates that treatments aimed at altering the femoral articulating surfaces may adversely affect one PF compartment for the sake of the other.

Muscle force and activation

Longitudinal studies examining the influence of muscle strength and activation on the progression of PF OA have focussed on the quadriceps ( Table 1 ). Quadriceps activity can promote cartilage health and stabilise the knee joint, however it may also increase the joint reaction forces . Over time, individuals with PFPS may modify their walking behaviour in an attempt to reduce this force. This is particularly the case for vastus medialis, which can be inhibited by knee pain Ref. . Two studies have evaluated the influence of quadriceps strength on PF structural progression and pain over 15–30 months. Neither study found a linear association between strength and joint space narrowing or cartilage loss ( Fig. 5 ). However, Amin et al. , observed the proportion of cartilage loss, pain and disability were less when quadriceps strength was high. Participants in both studies exhibited either PF or TF OA, thus results may over or under represent the influence of quadriceps strength for individuals with PF OA. In addition, the apparent protective effect was no longer evident when Amin et al. stratified participants based on the presence of varus alignment.

Lower limb alignment and kinematics

Over 18 months, valgus alignment resulted in a 1.6-fold increase in the odds of isolated lateral compartment progression ( Fig. 5 ). Varus alignment was associated with a two-fold increase in the odds of medial PF OA compartment progression. Further, as the severity of varus alignment increased, the odds of medial PF progression also increased. A potential explanation for the correlation between alignment and structural damage is that joint laxity increases with increasing malalignment, leading to greater shear stress a the PF joint. Sharma et al. explored this explanation in their longitudinal examination of TF OA progression. These authors hypothesised that if laxity influences OA progression, controlling for it during the analysis should decrease the strength of the alignment-progression relationship. This did not occur, suggesting that laxity may not be a mechanism for the effect of varus alignment on TF OA progression. While the generalizability of these results to PF OA is questionable, it suggests that other alignment and kinematic factors may influence knee OA progression. However, the influence of other potentially contributing factors, such as tibiofemoral rotation or patella kinematics, has not been longitudinally examined.