Knee ligament injury is one of the most frequent sport injuries and ligament reconstruction has been used to restore the structural stability of the joint. Cycling exercises have been shown to be safe for anterior cruciate ligament (ACL) reconstruction and are thus often prescribed in the rehabilitation of patients after ligament reconstruction. However, whether it is safe for posterior cruciate ligament (PCL) reconstruction remains unclear. Considering the structural roles of the PCL, backward cycling may be more suitable for rehabilitation in PCL reconstruction. However, no study has documented the differences in the effects on the knee kinematics between forward and backward pedaling. Therefore, the current study aimed to measure and compare the arthrokinematics of the tibiofemoral joint between forward and backward pedaling using a biplane fluoroscope-to- computed tomography (CT) registration method.

Eight healthy young adults participated in the current study with informed written consent. Each subject performed forward and backward pedaling with an average resistance of 20 Nm, while the motion of the left knee was monitored simultaneously by a biplane fluoroscope (ALLURA XPER FD, Philips) at 30 fps and a 14-camera stereophotogrammetry system (Vicon, OMG, UK) at 120 Hz. Before the motion experiment, the knee was CT and magnetic resonance scanned, which enabled the reconstruction of the bones and articular cartilage. The bone models were registered to the fluoroscopic images using a volumetric model-based fluoroscopy-to-CT registration method, giving the 3-D poses of the bones. The bone poses were then used to calculate the rigid-body kinematics of the joint and the arthrokinematics of the articular cartilage. In this study, the top dead center of the crank was defined as 0° so forward pedaling sequence would begin from 0° to 360°.

Compared with forward pedaling, for crank angles from 0° to 180°, backward pedaling showed significantly more tibial external rotation. Moreover, both the joint center and contact positions in the lateral compartment were more anterior while the contact positions in the medial compartment was more posterior, during backward pedaling. For crank angles from 180° to 360°, the above-observed phenomena were generally reversed, except for the anterior-posterior component of the contact positions in the medial compartment.

Forward and backward pedaling displayed significant differences in the internal/external rotations while the rotations in the sagittal and frontal planes were similar. Compared with forward cycling, the greater tibial external rotation for crank angles from 0° to 180° during backward pedaling appeared to be the main reason for the more anterior contact positions in the lateral compartment and more posterior contact positions in the medial compartment.

Even though knee angular motions during forward and backward pedaling were largely similar in the sagittal and frontal planes, significant differences existed in the other components with different contact patterns. The current results suggest that different pedaling direction may be used in rehabilitation programs for better treatment outcome in future clinical applications.