The purpose of the study was to identify risk factors that are associated with re-displacement of the hip after surgical reconstruction in cerebral palsy.

Retrospective review of children with cerebral palsy who had hip reconstruction with proximal femoral varus derotation osteotomy (VDRO) and Dega-type pelvic osteotomy, between 2005–2012, at a UK and European institution, was performed. Patient demographics, GMFCS, clinical and radiological outcome were assessed as well as the presence of pelvic obliquity and significant scoliosis (Cobb angle > 10 degrees). Redisplacement was defined as Reimer's Migration Index (MI) >30% at final follow-up. Logistic regression analysis was used to assess which factors were predictive of redisplacement and adjusted for clustered variables (α = 0.05).

Eighty hips were identified in 61 patients. The mean age at surgery was 8.8 years (± 3.3). Mean MI pre-op was 68% (± 23%) and post-op was 8% (± 12%). At a mean follow-up, of 3.2 years (± 2.0), 23 hips had a MI >30%. Of these; five were symptomatic, and one had required a salvage procedure. Metalwork removal was undertaken in 14 hips. Logistic regression demonstrated that the pre-operative MI and the percentage of acute correction were significant predictors of re-displacement. If the pre-operative MI was greater than 65 percent, the odds ratio (OR) for redisplacement was 5.99 (p = 0.04). If correction of the MI was less than 90% of the pre-operative MI, the OR for re-displacement was 4.6 (p = 0.03). Age at the time of surgery, GMFCS, pelvic obliquity and scoliosis were not predictive of re-displacement.

These results, firstly, highlight the importance of hip surveillance in children with cerebral palsy to allow timely intervention to ensure adequate radiological outcomes. Secondly, as in developmental hip dysplasia, full concentric reduction is essential to reduce the risk of re-displacement, with its associated clinical consequences.