Accuracy of patellar tendon at the attachment as anatomic landmark for rotational alignment of tibial component

Abstract

To investigate the accuracy of the modified Akagi line which referenced the patellar tendon at the attachment and the geometrical center point of the tibial osteotomy surface for tibial rotational alignment. Between July 2021 and December 2021, 72 patients who underwent three-dimension (3D) CT for varus osteoarthritis knees were enrolled. Among 72 patients, 18 were male and 54 were female with a mean age of 64.9 years (range, 47-84 years). The preoperative hip-knee-ankle angle ranged from 0° to 26°, with a mean of 9.3°. CT images were imported into Mimics 21.0 medical image control system to establish 3D models of the knees. The prominent point of lateral epicondyle and the medial epicondylar sulcus were identified in femoral 3D models to construct the surgical transepicondylar axis and the vertical line of its projection [anteroposterior (AP) axis]. In tibial 3D models, the patellar tendon at the attachment was used as anatomical landmarks to construct rotational alignment for tibial component, including the line connecting the medial border of the patellar tendon at the attachment (C) and the middle (O) of the posterior cruciate ligament insertion (Akagi line), the line connecting the point C and the geometric center (GC) of the tibial osteotomy plane [medial border axis of the patellar tendon (MBPT)], the line connecting the medial sixth point of the patellar tendon at the attachment and the point GC [medial sixth axis of the patellar tendon (MSPT)], the line connecting the medial third point of the patellar tendon at the attachment and point O [medial third axis of the patellar tendon 1 (MTPT1)], and the line connecting the medial third point of the patellar tendon at the attachment and point GC [medial third axis of the patellar tendon 2 (MTPT2)]. The angles between the five reference axes and the AP axis were measured, and the distribution of the rotational mismatch angles with the AP axis was counted (≤3°, 3°-5°, 5°-10°, and >10°). Relative to the AP axis, the Akagi line and MBPT were internally rotated (1.6±5.9)° and (2.4±6.9)°, respectively, while MSPT, MTPT1, and MTPT2 were externally rotated (5.4±6.6)°, (7.0±5.8)°, and (11.9±6.6)°, respectively. There were significant differences in the rotational mismatch angle and its distribution between reference axes and the AP axis ( F=68.937, P<0.001; χ 2=248.144, P<0.001). The difference between Akagi line and MBPT showed no significant difference ( P=0.067), and the differences between Akagi line and MSPT, MTPT1, MTPT2 were significant ( P<0.012 5). When the position of the posterior cruciate ligament insertion can not be accurately identified on total knee arthroplasty, MBPT can be used as the modified Akagi line in reference to the geometrical center point of the tibial osteotomy surface to construct a reliable rotational alignment of the tibial component.