A most important finding in this study was that the ACL was shorter during cross-leg motion than during squatting in mid-flexion. This fact suggested that the ACL is looser during cross-leg motion than during squatting. In other words, the ACL might be easy to affect by differences in flexion motions. Hence, ACL-preserved TKA might be able to reproduce the ACL length difference between squatting and cross-legged sitting. On the other hand, regarding length changes of the PCL, MCL, and LCL, there were no significant differences among the 3 motions. This suggested that the length changes of PCL, MCL, and LCL do not change even though the high-flexion motions are different. In addition, the ACL might be the most easily affected ligament of the knee in mid-flexion. The previous study demonstrated that the effect of coronal plane knee motion on cruciate ligament was larger than that of collateral ligament [32]. Additionally, ACL during Closed-Kinetic-Chain activities largely elongated, while that during Open-Kinetic-Chain activities slightly elongated [33]. Therefore, the ACLs have changed in cross-legged motion. In high flexion, the length change of the ACL was also not significantly different among the 3 motions. This suggested that any length changes of the ligaments do not change even though the flexion motions are different in high flexion. Therefore, regarding the ligament balance in bi-cruciate retaining TKA, the difference between the three activities might be low in high flexion.
From early flexion to mid-flexion, the ACL shortened with flexion. On the other hand, it extended slightly during high flexion. These tendencies were similar among the 3 motions. Additionally, the length change pattern with flexion was similar to that seen in previous in vivo studies using transducers [2, 3]. However, the length change of previous studies was less than 10%, while the length changes in the present study were more than 30%. Many factors such as ethnicity, gender, age, body mass index, or method of analysis contribute to the difference. Especially the method of analysis, the previous studies were analyzed from 0° to 90° of flexion. While this study was evaluated from 0° to 150° of flexion. Moreover, another previous study that evaluated the ligament length change using similar methods reported the ACL elongated more than 30% from 120° to 0° of flexion [14]. Hence, it was thought that the data in the current study was appropriate.
The PCL extended with flexion in the present study. Previous studies that investigated in vivo length change of the PCL using static methods also indicated the same pattern, including the absolute values [4, 34, 35]. This fact suggested that the PCL is commonly affected in high flexion, and the length change during dynamic motion was similar to that during static motion.
Regarding the MCL, adMCL, mdMCL, and asMCL extended from early flexion to mid-flexion. This trend was similar to previous studies [8, 10]. These facts suggested that the anterior portion of the MCL is easily affected in mid-flexion. Furthermore, the deep layer of the MCL might be easily affected in mid-flexion, because two thirds of the portion extended. During TKA for varus knee, we usually release the MCL to modify the soft tissue balance. Releasing the anterior and deeper layers of the MCL might be effective to modify mid-flexion balance.
Regarding the LCL, the anterior portion extended with flexion. This trend was similar to the previous study [10]. These facts suggested that the anterior portion could be affected in the lateral soft tissue balance, especially in high flexion. The high SD might mean high variation. Therefore, it might be impossible to detect the difference statistically because of the high variation. While this suggested that there is a high variation of LCL length change regardless of the type of high flexion activities in normal knees.
The high SDs in this study suggested many inter-subject variations in normal knees. However, the trends were similar to previous studies [8, 10, 36]. Therefore, it is thought that the results of this study were appropriate.
Several limitations of this study need to be discussed. First, the present study only analyzed Japanese male subjects. Female subjects or other races might display different length changes. Second, the number of volunteers involved was small. Therefore, the present results might not be generalizable to the general population. Third, only normal knees were evaluated. The length changes to ligaments of osteoarthritis knees and knees after TKA might be different from those of normal knees. Therefore, we will investigate the length changes to ligaments of their knees in our next study. Forth, single bundles of ACL and PCL were analyzed in the current study. Double bundles analysis might show different results. Fifth, the length of the direct line between the attachment areas was defined as the ligament length in this study. While, some studies used shortest three-dimensional wrapping path because the MCL wraps along the surfaces of the tibia and femur [8, 10]. Therefore, the ligament length of MCL might be shorter than that of the previous studies [8, 10]. However, a previous computer simulation study that used the straight line for ligament reported the ligament force was similar to the real ligament [37]. In addition, the ligament length change during squatting was similar to the previous studies [8, 10]. Therefore, it is thought that the length change of MCL in this study is appropriate. Sixth, although each volunteer practiced several times before the fluoroscopic analysis, the kinematics was measured only one time per activity due to the radiation dose limit. An intra-subject variation may also affect the results.
In conclusion, the ACL was shorter during cross-leg motion than during squatting in mid-flexion. This suggests that the ACL is looser during cross-leg motion than during squatting. On the other hand, the length changes of the PCL, MCL, and LCL did not change even though the high-flexion motions were different.