2015 ISAKOS Biennial Congress ePoster #1009
Pre-Operative Assessment of Femoral Rotation and its Relationship with Coronal Alignment: An MRI Study
Catherine Jane McDougall, MBBS, FRACS(orth), Brisbane, QLD AUSTRALIA
Price Gallie, MBBS, FRACS, FAOA(Orth), Southport, QLD AUSTRALIA
Sarah Whitehouse, PhD, Brisbane AUSTRALIA
Gold Coast University Hospital, Gold Coast, QLD, AUSTRALIA
FDA Status Not Applicable
Summary: This study uses axial MRI to define the rotational anatomy of the distal femur in the osteoarthritic knee and investigate its relationship with coronal alignment. We conclude that individual variability does exist between the commonly used rotational axes of the distal femur and that there is a difference between rotational alignment and changing coronal alignment in the osteoarthritic knee.
Obtaining correct rotational alignment in the femur is crucial for the success of total knee arthroplasty. The angular relationships between axes in the axial plane contribute to the positioning of the femoral component. Individual variation in axes is a potential source of error and previous CT studies have suggested that this variation is linked to overall coronal alignment.
The aim of this study is to use MRI to define the rotational anatomy of the distal femur in the osteoarthritic knee and investigate its relationship with coronal alignment.
We reviewed the pre-operative MRI scans of 126 patients with knee osteoarthritis. After determining the overall coronal alignment from lower limb alignment films, three groups were established, excessive varus ( >6°), neutral (>6° - < -6°), and excessive valgus ( <-6°).
On the axial image, two investigators marked the Posterior Condylar Line (PCL), Anatomical Epicondylar axis (AEA), Surgical Epicondylar Axis ( SEA) and AP Trochlea Axis (APA) using a 3D method with the Osirix Imaging Program. The relationship between each of the axes to the PCL was recorded. We assessed interobserver reliability, the overall mean for each of the axes and a comparison of the axes between the three alignment groups. Statistical analysis was performed using ICC, ANOVA and multicomparison testing.
There was excellent interobserver reliability for SEA and AEA. Compared to the Posterior Condylar Line, the mean SEA was 1.7 (1.3 to 2.03), AEA 5.6 (5.17 to 5.93) and APA 94.3 (93.8 to 94.8). There were statistically significant differences between the excessive varus and excessive valgus groups for SEA (0.91: 3.01 p<0.001) and AEA (4.68: 6.96 p <0.001). There was no statistical difference for APA (93.9: 95.3 p=0.238).
Individual variability does exist between the commonly used rotational axes of the distal femur. There is a statistically significant difference between rotational alignment and changing coronal alignment in the osteoarthritic knee. Better understanding of these relationships and appreciation of the variability may assist in better overall rotational alignment in total knee arthroplasty.