Long-term changes in femoral anteversion and hip rotation following femoral derotational osteotomy in children with cerebral palsy Elizabeth Boyer a , Tom F. Novacheck a,b , Adam Rozumalski a , Michael H. Schwartz, Ph.D. a,b, * a Gillette Children’s Specialty Healthcare, Center for Gait and Motion Analysis, St. Paul, USA b University of Minnesota, Department of Orthopaedic Surgery, Minneapolis, USA A R T I C L E I N F O Article history: Received 15 June 2016 Received in revised form 25 August 2016 Accepted 6 September 2016 Keywords: Cerebral palsy Femoral derotational osteotomy Torsion Gait analysis Outcome Recurrence A B S T R A C T Background: Excessive femoral anteversion is common in cerebral palsy (CP), is often associated with internal hip rotation during gait, and is frequently treated with a femoral derotational osteotomy (FDO). Concerns exist regarding long-term maintenance of surgical outcomes. Past studies report varying rates of recurrence, but none have employed a control group. Methods: We conducted a retrospective analysis examining long-term (5 years) changes in anteversion and hip rotation following FDO in children with CP. We included a control group that was matched for age and exhibited excessive anteversion (>30  ) but did not undergo an FDO. Anteversion, mean stance hip rotation, and rates of problematic remodeling and recurrence were assessed (>15  change and final level outside of normal limits). Results: The control group was reasonably well matched, but exhibited 9  less anteversion and 3  less internal hip rotation at the pre time point. At a five year follow-up, the FDO group had less anteversion than the control group (20  vs. 35  , p < 0.05). The mean stance phase hip rotation did not differ between the groups (4  vs. 5  , p = 0.17). Over one third of limbs remained excessively internal in both groups (FDO: 34%, Control: 37%). Rates of problematic recurrence and remodeling were low (0%–11%). Conclusions: An FDO is an effective way to correct anteversion in children with CP. Long-term hip rotation is not fully corrected by the procedure, and is not superior to a reasonably well matched control group. Rates of problematic recurrence and remodeling are low, and do not differ between the groups. ã 2016 Elsevier B.V. All rights reserved. 1. Introduction Excessive femoral anteversion is a common bony torsion present in cerebral palsy (CP) [1–4]. The natural history of anteversion remodeling in typically developing children is a decrease from approximately 40  at birth to 15  at skeletal maturity [3]. The etiology of excessive anteversion is speculative, but delayed or attenuated remodeling is the primary suspect. This hypothesis is supported by cross-sectional data; though longitu- dinal studies are lacking [1]. Skeletal loading plays a causal role in femoral morphology remodeling, and may result in atypical anteversion [5]. Greater anteversion is observed in the more- affected side of individuals with hemiplegia, again implicating muscle tone, neurological control, and skeletal loading [4]. Excessive anteversion in CP is often considered problematic for various reasons. If not accompanied by internal hip rotation, it may decrease coronal plane hip abductor moment arms (lever arm dysfunction), which may result in insufficient hip abductor moment generation, and thus lead to gait pathologies [6–8]. Internal hip rotation and concomitant in-toeing is also cosmeti- cally undesirable and generally follows excessive anteversion. In- toeing may also increase the risk of trips and falls, though no clear evidence exists to confirm this. The accepted treatment to address excessive anteversion and its associated gait problems is a femoral derotational osteotomy (FDO). Many studies have reported positive short-term improve- ments in anteversion and hip rotation (Table 1). McMulkin compared two groups of CP patients who underwent surgery, either with or without an FDO, and noted that the FDO group had favorable improvements in hip rotation during stance while the other group did not [9]. However, several studies have also reported instances of over- or under-correction of hip rotation. For instance, an FDO performed on a limb with mean stance hip rotation less than 15  may lead to an external hip rotation and out- toeing [10,11]. Long-term evaluations of FDO outcomes show varying levels of surgical correction maintenance (Table 1). Recurrence has been * Corresponding author at: Gillette Children’s Specialty Healthcare Center for Gait and Motion Analysis 200 University Avenue East St. Paul, MN 55101, USA. E-mail address: schwa021@umn.edu (M.H. Schwartz). http://dx.doi.org/10.1016/j.gaitpost.2016.09.004 0966-6362/ã 2016 Elsevier B.V. All rights reserved. Gait & Posture 50 (2016) 223–228 Contents lists available at ScienceDirect Gait & Posture journal homepage: www.else vie r.com/locate /gait post