Femoral Head Deformity Associated With Hip Displacement in Nonambulatory Cerebral Palsy: Results at Skeletal Maturity.

ABSTRACT

BACKGROUND: Maintaining femoral head shape (FHS) and acetabular sphericity are important goals in preventing long-term osteoarthritis in hips in children with cerebral palsy (CP). As acetabular morphology has been widely studied, our objective was to determine FHS in CP after triradiate cartilage (TRC) closure, a proxy for skeletal maturity, and the risk factors associated with residual deformity and osteoarthritis. METHODS: In this retrospective cohort study, patients with CP [Gross Motor Function Classification System (GMFCS) IV to V], minimum 4 yearly hip radiographs after age 10 years, and at least 1 radiograph after age 16 years, were included. Primary outcome was FHS (Rutz), stratified as "less severe" (Rutz A to B) and "more severe" (Rutz C to D). Secondary outcomes included migration percentage (MP), age at TRC closure, previous reconstructive (femoral with/without pelvic osteotomies) surgery, previous intrathecal baclofen, Tönnis osteoarthritis grade, and GMFCS level. Statistical analyses included χ 2 analysis and multiple logistic regression. RESULTS: One hundred sixty-three patients (326 hips) met the inclusion criteria, with TRC closure at age 14.0 (SD 1.8) years. At final follow-up of 4.4 (SD 2.4) years after TRC closure, 17% (55 hips), had a "more severe" FHS. From TRC closure to final follow-up, the frequencies of "less severe" hips decreased (-10%, P <0.001), while "more severe" increased (+115%, P <0.001). In multiple regression analysis, MP at TRC closure was the only significant risk factor associated with a "more severe" FHS at final follow-up ( P =0.03). Receiver operating characteristic curve analysis determined MP≥30.5% to be associated with a "more severe" FHS at final follow-up ( P <0.009). The FHS was not affected by reconstructive surgery, sex, GMFCS level, or intrathecal baclofen use. "Less severe" hips had lower Tönnis grades (0 to 1) compared with "more severe" hips (Tönnis grades 2 to 3) at final follow-up ( P <0.001). CONCLUSIONS: FHS at skeletal maturity was not influenced by prior reconstructive surgery but was negatively affected when MP≥30.5% at the time of TRC closure. The extent of residual femoral head deformity correlated with the severity of osteoarthritis at final follow-up. LEVEL OF EVIDENCE Level III.