Do we get a "real" alignment of knee in the preoperative planning of high tibia osteotomy: a prospective study of reproducibility.

BACKGROUND Precise preoperative measurement of knee alignment is needed to calculate the accurate angle of correction at proximal tibia osteotomy for medial gonarthrosis. METHODS We performed a prospective study to compare the reproducibility of measuring the mechanical and anatomical axes. Thirty-two patients (32 knees) with medial compartment osteoarthritis to be treated with proximal tibia osteotomy were included in this study. Preoperatively, whole lower limb roentgenographs were obtained twice, and 2 independent radiologists measured the mechanical and anatomical axes from each roentgenograph. RESULTS Measurement of mechanical and anatomical axes had a mean variability of 2.22 and 1.88 degrees, respectively, which was not statistically significant (p = 0.267) in the assessment of reproducibility. With the anatomic axis, however, we found 0.61 degree of variability to the roentgenographic procedures and 1.30 degree to the radiologists (p = 0.007). With the mechanical axis, the corresponding findings were 1.30 degree and 1.02 degree (p = 0.167). Despite the relative small number of patients in this series, errors in measurement of the anatomical axis seem mostly to originate from different radiologists, whereas errors in measurement of the mechanical axis seem to originate from both the radiologists and the procedures. The maximum variability in measuring both axes was 3 degrees, which is highly significant for a reliable calculation of the wedge when performing proximal tibia osteotomy. CONCLUSIONS We suggest that, for accuracy and economy, measurement of the anatomical axis might be better. Furthermore, by measuring either mechanical or anatomical axis, the errors originating from roentgenographic measurement of knee alignment should be considered in preoperative planning.