In Vivo Three-dimensional Kinematics of Normal Knees During Sitting Sideways

Background The normal knee kinematics during asymmetrical kneeling such as the sitting sideways remains unknown. This study aimed to clarify in vivo kinematics during sitting sideways of normal knees. Twelve knees from six volunteers were examined. Under uoroscopy, each volunteer performed a sitting sideways. A two-dimensional/three-dimensional registration technique was used. The rotation angle, varus-valgus angle, anteroposterior translation of the medial and lateral sides of the femur relative to the tibia, and kinematic pathway in each exion angle was evaluated.


Results
Bilateral knees during sitting sideways showed a femoral external rotation relative to the tibia with exion. Whereas the ipsilateral knees showed valgus movement, and the contralateral knees showed varus movement. The medial side of the contralateral knees was more posteriorly located than that of the ipsilateral knees beyond 110° of exion. The lateral side of the contralateral knees was more anteriorly located than that of the ipsilateral knees from 120° to 150° of exion. In the ipsilateral knees, a medial pivot pattern followed by a bicondylar rollback was observed. In the contralateral knees, no signi cant movement followed by a bicondylar rollback was observed.

Conclusion
Even though the asymmetrical kneeling such as sitting sideways, the knees did not display asymmetrical movement.
Kneeling is one of the most common activities of daily living such as sitting on the oor, gardening, and praying. Concerning the kinematics, several studies demonstrated that a femoral external rotation with a medial pivot is observed with the knee exion during kneeling [3,4]. There are many kinds of sitting styles such as sitting cross-legged, seiza-sitting, and sitting sideways. In addition, the kinematics during sitting cross-legged is different from that of kneeling and squatting [4]. However, most of the sitting styles evaluated previously were symmetrical [2][3][4]8]. Therefore, the kinematics during asymmetrical sitting such as the sitting sideways remains unknown.
Especially in Asian and Middle-Eastern countries, the patients after total knee arthroplasty (TKA) desire the sitting with deep-knee-bend. Several studies demonstrated that the knees after TKA during kneeling were also indicated the femoral external rotation [9][10][11]. Niki et al. reported that although the seiza-sitting after TKA seemed safe in terms of component dislocation [10]. Furthermore, Nakamura et al. reported that the ball-and-socket joint articulation enabled the patients to kneel safely without the incidence of any dislocation [11]. On the other hand, recently, medial pivot TKA was launched with to recreate normal knee kinematics. Additionally, many studies reported the clinical outcome was excellent or good [12][13][14][15][16][17]. In particular, for the patients after medial pivot TKA, the asymmetrical sitting such as sitting sideways tend to be avoided because the contralateral knee is capable to show the lateral pivot motion with the femoral internal rotation. However, whether the contralateral knee during sitting sideways would show the lateral pivot motion with the femoral internal rotation remain unknown. Therefore it is meaningful to evaluate the normal knee kinematics during sitting sideways.
This study aimed to clarify in vivo kinematics during sitting sideways of normal knees. The hypothesis of this study was that in vivo kinematics during sitting sideways was different between the ipsilateral knees and contralateral knees, in other words, the ipsilateral knees showed femoral external rotated, valgus, and medial pivot motion with exion, on the other hand, the contralateral knees showed femoral internal rotated, varus, and lateral pivot motion with exion.

Methods
Twelve knees from six volunteers were examined. All of the volunteers were Japanese males and provided written informed consent to participate in this study. Institutional review board approval was obtained, with documentation. At the time of examination, the mean age was 37.3 ± 7.6 years. The mean body height was 169.9 ± 5.2 cm. The mean body weight was 64.2 ± 5.2 kg. All of the values were expressed as mean ± standard deviation (SD).
Under uoroscopy, each volunteer performed a sitting sideways at a natural pace ( Figure 1). They practised the motion several times before recording. The right and left knee motion were separately recorded. The sequential motion was recorded as a series of digital X-ray images (1024 × 1024 × 12 bits/pixel, 7.5-Hz serial spot images as a DICOM le) using a 17-inch (43-cm) at panel detector system. Furthermore, all images were processed by dynamic range compression, thereby enabling edge-enhanced images. To estimate the spatial position and orientation of the knee, a two-dimensional/threedimensional (2D/3D) registration technique [4, 18] was employed.
Three-D bone models were created from computed tomography (CT) and used for the computer-aided design (CAD) models. Estimation accuracy for the relative motion between 3D bone models was ≤ 1° in rotation and ≤ 1 mm in translation [4]. A local coordinate system at the bone model was produced according to a previous study [19]. Knee rotations were described using the joint rotational convention of Grood and Suntay [20]. Femoral rotation angle relative to the tibia, varus-valgus angle, and anteroposterior (AP) translation of the medial sulcus (medial side) and lateral epicondyle (lateral side) of the femur on the plane that is perpendicular to the tibial mechanical axis in each exion angle were evaluated [4]. AP translation was calculated as a percentage relative to the proximal AP dimension of the tibia [4]. External rotation was denoted as positive and internal rotation as negative. Valgus was denoted as positive and varus as negative. Positive and negative values of AP translation were described as anterior and posterior to the axis of the tibia, respectively.

Statistical analyses
Results were analysed using SPSS version 24 (IBM Corp., Armonk, NY, USA). Repeated measure analysis of variance (ANOVA) and post hoc pairwise comparison (Bonferroni test) was used to analyse all evaluation items. A p-value <0.05 was considered statistically signi cant. Moreover, a power analysis using EZR [21] indicated that 11 knees would be required for an alpha set at 0.05 and power at 0.8.

AP translation
The AP translation of the medial side of the ipsilateral femur indicated 10.4 ± 7.0 % posterior movement from 130° to 150° of exion. The AP translation of the medial side of the contralateral femur indicated 11.0 ± 8.6 % posterior movement from 130° to 150° of exion (Figure 4). Beyond 110° of exion, the medial side of the contralateral knees was signi cantly more posteriorly located than that of the ipsilateral knees (p = 0.011).
Kinematic pathway (Figure 6) In the ipsilateral knees, the difference between the medial and lateral sides of the femur re ected a medial pivot pattern from 110° to 130° of exion. From 130° to 150° of exion, a bicondylar rollback was observed. In the contralateral knees, no signi cant movement was observed from 110° to 130° of exion. From 130° to 150° of exion, a bicondylar rollback was observed.

Discussion
This is the rst study to evaluate the in vivo knee kinematics of normal volunteers during asymmetrical kneeling using the CAD model of uoroscopically captured images. Regarding the varus-valgus angle, the ipsilateral knees during sitting sideways showed valgus movement, on the other hand, the contralateral knees during sitting sideways showed varus movement. This fact was as hypothesized. Whereas, regarding rotation angle and AP translation, the result of this study was contrary to the hypothesis. In other words, the bilateral knees showed the femoral external rotation, additionally, the contralateral knees did not show a lateral pivot motion. These facts suggest that regardless of the varus-valgus movement, normal knees during asymmetrical kneeling display a femoral external rotation. Moreover, a lateral pivot motion was not observed even though the contralateral knees during asymmetrical kneeling. Namely, even though the asymmetrical kneeling, the knees did not display asymmetrical movement. However, the extent of rotation angle and AP translation was signi cantly different between the ipsilateral knees and contralateral knees; the femoral external rotation of contralateral knees was smaller than that of ipsilateral knees, the medial side of contralateral knees was located more posterior than that of ipsilateral knees, and the lateral side of contralateral knees was located more anterior than that of ipsilateral knees. Therefore, although the kinematics of contralateral knees during sitting sideways was not contrasted with that of ipsilateral knees perfectly during sitting sideways, it might be more laterally constrained movement.
Murakami et al reported that the knees during an asymmetrical activity such as golf swing displayed asymmetrical movement, unlike this study. The golf swing is a closed-kinetic-chain activity, whereas the sitting sideways is an open-kinetic-chain activity. This suggests that even though the asymmetrical activities, the kinematics is different depending on each activity. Moreover, in sitting sideways, the respective movement of femur and tibia may be different between the ipsilateral knees and contralateral knees.
The previous study that evaluated the knee kinematics during symmetrical kneeling has reported that the range of femoral external rotation with exion was 14.8 ± 3.8°, and the range of lateral AP translation with exion was 40.2 ± 10.2 % [4]. The rotation and lateral AP translation of ipsilateral knees during sitting sideways were similar to that of normal knees during symmetrical kneeling [3,4]. On the other hand, the rotation and lateral AP translation of contralateral knees during sitting sideways were smaller than those of normal knees during symmetrical kneeling [3,4]. These suggest that patients who underwent TKAs do not need to avoid sitting sideways. However, in particular, those who underwent guided-motion TKAs which guide medial pivot motion should take care of the sitting sideways.
Regarding the kinematic pathway, a bicondylar rollback was observed from 130° to 150° of exion in both the ipsilateral and contralateral knees during sitting sideways. The previous studies demonstrated that normal knee kinematics showed a medial pivot pattern during kneeling [3,4]. These suggest that the normal knee kinematics during kneeling is different between symmetrical activity and asymmetrical activity in high-exion.
Gladnick et al reported that although from extension to 90° of exion normal knee exhibited an increase in varus-valgus laxity as the exion angle increase, the variability of AP translation in response to the varus and valgus load was small [22]. On the other hand, in the current study, the ipsilateral knees displayed valgus movement, whereas the contralateral knees displayed varus movement. Furthermore, the AP translation was different between the ipsilateral knees and contralateral knees. These suggest that during a high-exion activity more than 90° of exion, because the varus-valgus laxity increase additionally, the varus and valgus load may affect the AP translation.
This study has some limitations. First, this study analysed the knee joint kinematics of only Japanese males. The knee kinematics of females or the other races may be different. Second, in the current study, the right and left knee motion were separately recorded because it is impossible to record the bilateral knees in the at panel. Therefore, it is capable not to be re ected in the simultaneous knee motion during sitting sideways.

Conclusions
Even though the contralateral knee during sitting sideways, the femoral external rotation relative to the tibia was observed. Moreover, a lateral pivot motion was not observed. In other words, the knees during asymmetrical kneeling did not display asymmetrical movement.  study was approved by Osaka University ethics committee (Number 13106). All of the volunteers in this study provided written, informed consent prior to participation.

Consent for publication Not Applicable
Availability of data and materials The datasets analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was not supported by any funding.
Authors' contributions KK, TT, and SK designed the study and the initial draft of the manuscript. KK, TT, SK, and TY contributed to analysis of the kinematic data. HI, ST, ST and KS contributed to analysis a data and manuscript preparation. All other authors have contributed to data collection and interpretation, and critically reviewed the manuscript. The nal version of the manuscript was approved by all authors. Figure 1 Sitting sideways. The ipsilateral knee is indicated as the blue circled knee. The contralateral knee is indicated as the yellow circled knee.

Figure 2
Rotation angle during sitting sideways. The markers indicate the femoral rotation relative to the tibia. *, signi cant differences between ipsilateral knees and contralateral knees (p < 0.05) Figure 3 Varus-valgus angle during sitting sideways. The markers indicate the femoral varus-valgus movement relative to the tibia. *, signi cant differences between ipsilateral knees and contralateral knees (p < 0.05) Figure 4 Anteroposterior (AP) translation of the femoral medial sulcus during sitting sideways. AP translation was calculated as a percentage relative to the AP length of the tibia. *, signi cant differences between ipsilateral knees and contralateral knees (p < 0.05) Figure 5