Popliteal vascular safety during high tibial osteotomy : an analysis in 3- dimensional knee flexion models

Abstract

Concern about inadvertent popliteal vascular injury during high tibia osteotomy (HTO) for the posterior cortex of the proximal tibia requires careful attention. However, different methods and conditions adopted to each study led to draw conflicting conclusions. In addition, most studies performed in the context of two dimensions rather than three dimensions, and no studies have determined whether the results vary in relation to the real osteotomy plane. The objects of this study were as follows; (1) to develop a validated magnetic resonance imaging (MRI) based 3D knee flexion model that enables to evaluate knee and popliteal artery (PA) kinematics of each subject, (2) to analyze popliteal artery movement in the 3D coordinate system in relation to knee flexion and osteotomy techniques (lateral closed HTO (LCHTO) vs. uniplane medial open HTO (UP-MOHTO) vs. biplane medial open HTO (BP-MOHTO)), (3) to identify optimal techniques using oscillating saw or osteotome in each osteotomy plane. Sixteen subjects who underwent knee MRI scan with extension and 90° flexion in our institute were enrolled in the study. After developing the 3D knee flexion models for each specimen, 3 types of virtual osteotomies were performed in the models using avalidated sofware (Mimics® 17.0, Materialise, Leuven, Belgium). Displacement of the PA during knee flexion along the X-axis (dX) and Y-axis (dY), the distance between posterior tibial cortex and PA parallel to Y-axis (d-PCA). And also, ∆d-PCA was defined as the value obtained by substracting d-PCA of extension from d-PCA of flexion. Frontal plane safety proportion (FPSP) was define as the ratio of the length between the most medial or lateral margin and the medial or lateral edge of the PA to the total length of the osteotomy along the osteotomy plane. The maximal axial safe angles (MASA) of osteotomy calculated the angle was formed by two intersection line which lie in the osteotomy plane as well as pass through a common intersection point that is the the most medial or lateral eminence of the cutting surface. One line is parallel with the coronal plane of the coordinate system and the oother line is tangent to the surface of the popliteal artery along the osteotomy plane. Differences among 3 osteotomy methods were compared in each flexion angle. In LCHTO group, the mean dY, dX, ∆d-PCA were 1.9 ± 1.3 mm, 0.0 ± 5.7 mm, 1.3 ± 2.3 mm (p = 0.170), respectively. In UP-MOHTO group, the mean dY, dX, ∆d-PCA were 1.7 ± 1.7 mm, -0.1 ± 5.6 mm, 1.3 ± 1.8 mm (p = 0.050). In BP-MOHTO group, the mean dY, dX, ∆d-PCA were 1.8 ± 1.8 mm, -0.5 ± 5.7 mm, 1.7± 2.0 mm (p = 0.015). Regards to d-PCA, values in knee flexion was decreased rather than those in extended position, which were 6 subjects (37.5%) in LCHTO, 5 subjects (31.3%) in UP-MOHTO, 2 and 3 subjects (18.8%) in BP-MOHTO. The mean FPSP of each osteotomy method in knee extension were 37.6 ± 5.9 %, 49.2 ± 6.0 %, and 45.1 ± 8.1 % (LCHTO, UP- MOHTO, and BP-MOHTO, respectively). The mean MASA of each osteotomy method in knee extension were 45.8 ± 4.4°, 35.2 ± 5.9°, and 38.9 ± 6.5° (LCHTO, UP-MOHTO, and BP-MOHTO, respectively). In all ostetomy methods, values in knee flexion slightly increased, but these increments were not significnat. On average, the PA moved posteriorly during knee flexion based on reference points as locations of the PA in extension in all osteotomy planes. The d-PCA significantly increased toward lower level osteotomy planes. Although the significant increment of the mean value during knee flexion was noted in BP-MOHTO group only, these small (1.7 mm) increment and inconsistent movement along subjects may not provide clinical significance. In the frontal plane, the location of the PA was about 37% from most lateral cortex of the tibia, and 45 ~ 49 % from most medial cortex of the tibia along each osteotomy plane. Saw angle from the frontal plane should be at maximum 45° in the LCHTO and 35 ~ 40° in the MOHTO.