Introduction

Accurate implant placement in total knee arthroplasty (TKA) is one of the most important factors to achieve excellent results. There are many studies about coronal alignment of TKA but few about three-dimensionally evaluation of femur morphology as it pertains to navigation-aided surgery. The mechanical axis (MA) on coronal plane in the current navigation system is determined with a femoral head center and a distal reference point of the condylar center without considering femoral bowing. But recent studies show that distal femoral valgus angle is subject-dependent and variable of femoral anatomy. Thus, the concept of an ideal valgus angle of the femoral component with respect to MA and/or distal femoral axis is controversial. Information on the relationship among MA and femoral anatomical axis is important for preoperative and intraoperative evaluations for the navigation-aided TKA. Therefore, the objective of this study was to investigate the 3D coronal femoral alignment using a 3D-CT model of the whole femur.

Method

Consecutive 74 patients, including 140 femurs, who underwent whole leg CT preoperatively were used for this IRB-approved study. Patients with hip OA and previous intervention to the femur were excluded (n=3). This resulted in 59 females and 14 males whose mean age was 76.1 years old and mean BMI was 26.5 kg/m2. The 3D bone model was made from CT and was used to determine the following morphological parameters: the centroid of femoral head5 and distal end of femoral trochlea were calculated as FH and FT, respectively, and the line FH-FT was named the mechanical axis (MA). The distance between FH and FT was the femur length. A reference plane was defined as a plane including MA and parallel to a line connecting the most posterior points of the medial and lateral condyles as coronal plane. The centroids of cross sections below the lowest portion of the lesser trochanter (F1), 50 mm distal portion from F1 (F2), 110 mm proximal portion from FT (F3) and 60 mm proximal portion from FT (F4) were calculated. Lines F1-F2 and F3-F4 were named the proximal femoral axis and the distal femoral axis, respectively. The angles between proximal and distal femoral axes to MA on the reference plane were named the proximal femoral valgus angle (PFVA) and distal femoral valgus angle (DFVA), respectively. PFVA and DFVA were compared with a pared t-test. The gender effect of PFVA and DFVA was analyzed by an unpaired t-test. Correlations between PFVA or DFVA with height, weight, BMI and femur length were analyzed by linear regression analysis.

Results

The DFVA (8.0 ± 2.0°; mean ± SD) was greater than PFVA (3.2 ± 2.1°; p < 0.0001). While there was no significant difference in DFVA between female and male, PFVA in females (3.0 ± 2.1°) was smaller than in males (4.1 ± 1.7°; p < 0.02). The DFVA was greater than PFVA in females and males (p < 0001each) after split by genders.

Significance

The measurement system of the coronal alignment of the femur proposed in this study and preliminary values of the DFVA would be useful to design prostheses, surgical jigs, navigation systems and preoperative planning for TKA.