Introduction: It has been proposed that rotational gait abnormalities in the normal child are usually reflections of the anatomic deformity. A decreased acetabular and femoral anteversion have been recognized as a predisposing factor for osteoarthritis of the hip and the McKibbin instability index was introduced to quantify this relationship. Additionally, an increased femoral anteversion has been associated with osteoarthritis of the knee. However, it is well known that compensatory factors influence the dynamic rotational profile during gait. We compared rotational computed tomography data with gait analysis to evaluate their correlation and to elucidate the influence of compensatory mechanisms.

Materials and Methods: In a prospective study conducted between 2001 and 2005 patients presenting with rotational malalignment were sent for 3D gait analysis. Main exclusion criterion was any kind of neurological affection. Patients in whom surgery was considered were referred to rotational computed tomography. The rotational alignment of the pelvis, hip and knee at different times during the gait cycle as evaluated in the 3D gait analysis was compared to the angular values derived from the rotational computed tomography for the femur and tibia and statistically analyzed and correlated.

Results: There were 12 female and 16 male patients with a mean age of 16 (± 9.7) years at the time of gait analysis. After a first evaluation of data 8 limb segments were excluded to increase the quality of data. The mean anteversion of the femur was 29 degrees (2 degrees of retrotorsion to 56 degrees of anteversion) and the mean tibial torsion was 31 degrees (1 to 66 degrees of external torsion). The calculation of the Pearson correlation showed that an increase of femoral anteversion resulted in an increase of pelvic range of motion. An increase of femoral anteversion resulted in an increase of the internal rotation of the hip. Highly significant correlations were found between the rotational–CT values for the tibia and the all parameters describing rotation of the knee. The determination coefficient was high for tibial torsion versus knee rotation (R2 = 0.64), but showed a low value for femoral anteversion versus hip rotation (R2 = 0.2).

Conclusion: The rotation of the hip as found in the gait analysis showed only weak correlation with rotational CT data. This is not surprising as the hips segment offers many possibilities for compensation. The torsion of the tibia was found to correlate very strongly with the gait analysis. The McKibbin index seems questionable as a prognostic factor for the individual patient in the light of a multitude of dynamic compensatory influences. Effort should be made to integrate the static instability index with dynamic gait analysis data.