The association between tibial torsion, knee flexion excursion and foot progression during gait in people with knee osteoarthritis: a cross-sectional study

Osteoarthritis (OA) is a common chronic condition affecting more than 300 million people worldwide [1]. Knee OA is characterized by pain, knee joint stiffness, and limited knee joint range of motion. Common risk factors associated with knee OA include age, gender, malalignment, and other medical conditions [2, 3].

Altered gait kinematics at the knee joint and lower limb have been reported in individuals with knee OA. Specifically, individuals with knee OA have been observed to exhibit greater knee flexion at heel strike, reduced knee flexion excursion during early stance [4, 5], and altered foot progression angle [6, 7] during gait. Although research shows that altered sagittal plane knee kinematics [8] and foot progression angle [9] during gait was associated with joint loading, little is known on the mechanical factors that play a role in the development of these altered gait kinematics.

From a mechanical perspective, joint motion is determined by joint architecture. Malalignment of the lower limb might be an important factor that affects knee joint kinematics during gait. Tibial torsion, as a critical component of torsional alignment of the lower limb, was altered in those with knee OA. Smaller external tibial torsion was detected in people with knee OA than healthy controls [10]. Torsional malalignment of the lower limb is one of the factors in relation to abnormal gait rotational changes [11]. Knee joint has a locking mechanism: when it moves from 30° knee flexion to fully extension (0^°), a torsional motion is associated [12]. During the loading response of the stance phase, knee joint moves from extension to flexion with an associated external torsional motion on the tibia, which is known as the unscrew-home mechanism the knee joint [13]. This coupling mechanism is disrupted in people with medial knee OA [14]. It implies an interaction between torsional alignment and sagittal motion. Knee flexion–extension motion in early stance phase could be altered due to tibial torsion in people with knee OA. A computerized model study demonstrated that tibial torsion could affect activation of lower limb muscles and influence knee flexion–extension movement [15]. While studies have shown that tibial torsion can affect knee flexion motion, the exact nature of this relationship is not fully understood, especially in people with knee OA. External tibial torsion was also associated with abnormal foot progression angle in pediatric patients waiting for surgery [16]. Whether similar relationship happens in senior with knee OA has not been investigated. Such information is important because foot progression angle was associated with joint loading. The limited understanding of the complex biomechanical interactions between tibial torsion and lower limb kinematics during gait constrains the treatment, and prevention of related conditions in knee OA.

Aside from alignment, the presence of marginal osteophytes and joint space narrowing, which are the determinants of radiographic severity of knee OA, could restrict both passive and active knee range of motion [17]. Thus, the relationship between lower limb alignments and knee flexion motion during early stance phase of gait might be affected by these differences in bone and/or joint structures in people with different severities of knee OA. It is essential to explore the impact of disease severity since there is a possible effect of structural changes on gait kinematics. Findings from this study would enhance our understanding on the mechanical loading in people with knee OA.

Out of 100 potential participants with knee OA admitted between March 2017 and July 2018, forty-seven were recruited after screening for selection criteria. Most exclusion cases mainly had severe low back pain, recent knee injuries or surgeries, or the presence of lateral knee OA.

The present study aimed to explore the relationships between tibial torsion and lower limb kinematics at early stance phase during gait. Findings from this study provide evidence that lower limb alignment was associated with early stance knee flexion excursion and foot progression angle in people with medial knee OA. Specifically, tibial torsion was related to foot progression angle in all participants, and it was also associated with early stance phase knee flexion excursion in those with moderate knee OA. Knee varus angle showed a negative association with knee flexion excursion.

The novice finding from this study was a negative relationship between external tibial torsion angle and knee flexion excursion during early stance phase of gait in participants with moderate knee OA. Such finding indicated that in participants with moderate OA severity, smaller knee flexion excursion was related to greater external tibial torsion. This association between tibial torsion and knee flexion excursion would support our initial assumption that torsional alignments would possibly affect knee motion in sagittal plane. This result was consistent with a previous study based on a computerized model that tibial torsion could possibly influence the positioning of knee flexion–extension axis in the horizontal plane [15]. With more external rotated tibia, the more proximal parts of lower limb could have more internal rotation to attempting a more neutral lower limb alignment. In the loading response phase, knee joint flexes in a close chain manner from nearly full extension. The external tibial torsion could disrupt the unlocking mechanism of knee joint for initiating flexion which requires external rotation of femur on tibia in the loading response phase of gait cycle in moderate OA. Loss of such mechanism and having inadequate knee flexion is harmful to knee joint, since it had been discussed that larger knee flexion at early stance phase can facilitate impact absorption and loads transmission [25, 26]. In our previous report, we found that higher external tibial rotation was associated with greater external knee adduction moment in early stance phase in moderate knee OA, which indicated possible overload of knee joint medial compartment because of the high external tibial torsion [27]. When external tibial torsion was more likely associated with smaller knee flexion excursion, the external tibial torsion might be a potential harmful factor to knee OA mechanically. In the meanwhile, the present study found this relationship was specific to radiographic severity of OA. It agreed with our hypotheses that the relationship would be different in the two subgroups, and this difference is potentially related to progressive structural destruction of knee joint in these patients. In the present study, the relationship between tibial torsion and knee flexion could not be established in mild knee OA. Knee flexion excursion was smaller in people in early OA than established OA [28], and smaller in moderate OA than more severe OA [29, 30]. The present study detected an 80% smaller knee flexion excursion in the moderate than the mild knee OA group. Therefore, a decreasing trend of knee flexion excursion could be proposed with an advance in OA severity from the mild to moderate stage. People with moderate knee OA suffered more from stiffened knee sagittal motion. Nevertheless, there was no significant difference of tibial torsion angle between these two subgroups in the present study. Given the fact that the sample size was relatively small, the reason why only subgroup of moderate OA showed a negative relationship between knee flexion excursion and external tibial torsion remains unclear. We speculated that knee joint deformation could be the reason for this relationship. The severer osteophytes and in moderate knee OA could possibly hinder the coupling mechanism between torsional angle and knee flexion–extension motion. Nonetheless, the OA severity specific correlation hinted that tibial torsion could be more influential in the kinetic chain of lower limb in these more severe patients. The more proximal part in the kinetic chain, knee joint was less functioned and left less freedom for compensatory motion during gait.

Our study also detected a significant association between external foot progression angle and external tibial torsion. Thus, greater external tibial torsion occurred more possibly in people walking with greater external foot progression angle. Torsional alignments of the lower limb were closely interacted to maintain a relatively neutral position [31]. It had been suggested that tibial torsion was a strategy to neutralize abnormal foot progression angle [32]. In healthy population, the amount of tibial torsion angle was observed to have a positive association with foot progression angle [33]. Findings from this study showed a similar relationship in participants with knee OA. In adult population, the foot progression angle is external rotated for approximately 13° [34]. With this reference value, 43.3% of and 35.3% of our participants with the mild OA and moderate OA walked with greater external foot progression angle, respectively. The gait pattern with a large external foot progression angle was known as toe-out gait. The toe-out gait modification was associated with reduction in knee flexion excursion during the early stance phase of gait [35]. Gait re-education, such as an adoption of a toe-out gait, has been suggested to people with knee OA, because a toe-out gait had a favorable effect on de-loading the medical compartment of knee joint comparing to natural gait by reducing peak external adduction moment which serves as a surrogate for medial-to-lateral knee loading [36, 37]; however, the reduction existed at the second half of stance phase but not in the early stance phase, according to the summarized information from a meta-analysis [6]. Therefore, the beneficial effect of the toe-out gait on de-loading knee joint in loading response phase remained uncertain. On the other hand, findings from present study indicated that a larger external foot progression angle toe-out gait might be associated with external tibial rotation; and increase in external tibial rotation might relate to reduced knee flexion excursion during the early stance phase in those with moderate knee OA. The adoption of toe-out gait training might have different effects due to disease severity. Further research on this topic is warranted to confirm whether a disease severity specific gait modification is needed.

Our study also detected those participants with greater knee varus angle had smaller early stance flexion excursions. This was consistent with a previous report that showed the early stance phase knee flexion excursion in people with knee OA with varus malalignment was significantly smaller than that of healthy control or people with knee OA without varus malalignment [38]. In addition, there was a negative association between medial joint space narrowing of knee and early stance phase knee flexion excursion in people with medial knee OA [39]. Since knee varus malalignment could cause narrowing of medial joint space [40], it might affect the knee flexion excursion in the early stance phase.

This study was a cross-sectional study; therefore, the causal relationship could not be established according to the design and finding of current study. Also, our findings cannot be generalized to participants with severe knee OA, those with KL grade 4. One of the limitations is that potential confounders, such as knee joint contracture, tendon tightness, foot position, etc., were not measured. Further study could be designed to explore the change of foot progression angle on knee flexion excursion during gait in people with different OA severity.

We concluded that tibial torsion was related to early stance knee flexion excursion and foot progression angle during gait in people with medial knee OA. More specifically, external tibial torsion was associated with a smaller stance phase knee flexion excursion in moderate knee OA, and it was identified as a factor related to external foot progression angle. Such findings suggest that lower limb alignments in both horizontal and frontal planes would influence knee and foot kinematics during stance phase of gait. This OA severity specific relationship provided a new clue for future study to investigate a different gait modification strategy in people with moderate knee OA of excessive external tibial torsion.