Clinical Orthopaedics and Related Research ®

A Publication of The Association of Bone and Joint Surgeons ®

Determinants of Hip Displacement in Children With Cerebral Palsy

Chia Hsieh Chang MD, Ying Chih Wang MD, Pei Chi Ho MS, Ai Wen Hwang MD, Hsuan Kai Kao MD, Wei Chun Lee MD, Wen E. Yang MD, Ken N. Kuo MD



Coxa valga and femoral anteversion often are seen in patients with spastic hip displacement and osteotomy is recommended. However, the relationship between femoral deformities and hip displacement has not been clearly defined and other factors, such as joint motion and posture, should be considered before recommending treatment.


For children with cerebral palsy with Gross Motor Function Classification System Level IV or V, we questioned (1) whether hip abduction range correlates with hip displacement, (2) what the relationships are between proximal femoral deformities and hip displacement, and (3) whether the patient with a windblown posture has greater degrees of femoral anteversion?


We retrospectively studied 31 consecutive children with cerebral palsy with Level IV or V gross motor function who underwent three-dimensional CT for preoperative assessment of hip displacement between January 2010 and December 2013. Among the children, 23 had a windblown posture and eight had symmetric hip motion. Femoral anteversion and true neck-shaft angle were measured from the three-dimensional CT images. Migration percentage was the dependent variable we chose to study in relation to femoral anteversion, neck-shaft angle, maximal hip abduction, and hip flexion contracture, using correlations and multiple linear regressions. Using ANOVA and Scheffé’s post hoc tests, we analyzed and compared the data of 23 abducted hips and 23 adducted hips in the 23 children with windblown posture and in 16 displaced hips in the eight children with symmetric hip abduction.


Greater migration percentage was associated with less hip abduction range (r = −0.86; p < 0.001). Femoral anteversion had a weak correlation (r = 0.28; p < 0.05) to migration percentage, and the association became insignificant after considering hip abduction motion. Adducted windblown hips had greater femoral anteversion than the symmetric displaced hips and abducted windblown hips (46° vs 36° and 38°, respectively; p < 0.05).


Our study results did not support a relationship between femoral deformities and hip displacement after considering gross motor function and hip abduction motion. Greater femoral anteversion was noted in the adducted hips of patients with windblown posture, and derotation osteotomy is especially recommended.

Level of Evidence

Level III, diagnostic study.

Back to top