Radiographic Assessment of Coronal Alignment of the Lower Limb in End Stage Osteoarthritis of the Knee

Abstract

The position of the femoral component in the coronal plane is one of the determinants of postoperative alignment of the lower limb in the coronal plane which influences the survival of the prosthesis. Ideally the femoral component should be implanted perpendicular to the femoral mechanical axis. To do this, the valgus correction angle used to perform the distal femoral cut should be equal to the angle subtended by the mechanical and distal anatomical axes of the femur in the coronal plane. This angle can be accurately determined using preoperative whole lower limb (hip knee ankle) radiographs. Alternatively a presumptive distal femoral cut can be executed using a valgus correction angle of between 50 and 70. This assumes an ideal relationship between the mechanical and distal anatomical axes of the femur however this relationship is influenced by variations in femoral anatomy specifically the neck shaft angle and bowing of the femur in the coronal plane. Variation of these determinant factors between populations may make the use of presumptive cuts inaccurate resulting in incorrect positioning of the femoral component which contributes to post-operative malalignment of the limb in the coronal plane. These variations may explain the lack of consensus between authors on the safety of use of presumptive cuts and the need to perform routine preoperative whole lower limb radiographs to accurately determine the valgus correction angle to use when performing the distal femoral cut. The objective of this study was to establish the pattern of variation of the valgus correction angle as well as its determinants,coronal femoral bowing and neck shaft angle, in Kenyan patients with end stage osteoarthritis of the knee. The findings have implications for operative practice including the use of preoperative whole lower limb radiographs in routine total knee replacements as well as the choice of valgus correction angle to use in the event of performing presumptive distal femur cuts in the study population. Objective: To determine the pattern of variation of the valgus correction angle, bowing of the femur in the coronal plane and the neck shaft angle in patients with end stage osteoarthritis presenting in four orthopaedic centres for total knee replacement.

Design Cross-sectional observational study Setting The study was conducted at four orthopaedic centres namely Kenyatta national hospital, St. Francis community hospital, PCEA Kikuyu and the Aga Khan hospitals. Patients and Methods The patients were screened in the orthopaedic clinics of the respective hospitals and those with end stage osteoarthritis were selected for inclusion in the study. Patients were recruited until the target of 80 lower limbs with end stage osteoarthritis was reached. This resulted in the inclusion of 48 patients into the study. Data on the patient age, height, weight, time up and go and stair climbing test was collected. Weight bearing-whole lower limb radiographs were taken with each patient positioned with the lower limbs in 150 of internal rotation. Radiographic land marks described by Mullaji et al(1) were used to draw axes in the coronal plane and measurement of the angles of interest: the valgus correction angle, angle of coronal femoral bowing and the neck shaft angle were made. Data was keyed into the Statistical Package for Social Scientists and analysed. p values less than 0.05 were taken to be statistically significant. Results The valgus correction angle predicted by the whole lower limb radiographs averaged 6.290±1.800 for all limbs regardless of alignment and 6.50 ± 1.950 for patients with varus lower limb alignment. Of the lower limbs studied, 9.8% (8) had a predicted valgus correction angle greater than or equal to 90. Bowing of the femur averaged 2.450 ± 3.140 in all lower limbs regardless of alignment and 2.540 ± 3.390 in limbs with varus lower limb alignment. There was a strong positive correlation between bowing of the femur and the valgus correction angle r = 0.857 p < 0.01 for all lower limbs and r = 0.907 p < 0.01 for the limbs in varus alignment. Analysis of the impact of the neck shaft angle on the valgus correction angle in patients with varus lower limb

alignment demonstrated a significant difference between patients with coxa vara from those with coxa valga p < 0.05. Conclusions The average valgus correction angle in the population studied is significantly less than that reported by Mullaji et al ,7.30 ± 1.60, but significantly greater than that reported by Kharwadkar et al, 5.40 ± 0.90. The average bow was lower than that reported by Mullaji et al however this difference was not significant. The demonstration of a significantly lower valgus correction angle in individuals with a coxa valga compared to those with coxa vara corroborated the findings of Bardakos et al. These findings show that coronal femoral bowing and the neck shaft angle affect the relationship between the anatomical and mechanical axes of the femur and given their variation between populations may influence the accuracy of use of presumptive distal femoral cuts and therefore the need to take preoperative whole lower limb radiographs to precisely determine the valgus correction angle.