Intraoperative Hip Arthrography to Guide Decision-Making in Cerebral Palsy Hip Reconstruction

Background: Hip reconstruction surgery is indicated in nonambulatory children with cerebral palsy (CP). The decision to perform a pelvic osteotomy (PO) with a proximal femoral varus derotational osteotomy (VDRO) is subjective. Our hypothesis was that evaluation of dynamic stability and labral pathoanatomy using intraoperative hip arthrography (IHA) would reduce the performance of concomitant PO without adversely affecting outcomes. The capability of IHA to correctly guide intraoperative decision-making was evaluated using midterm outcomes in children with CP (CwCP) treated with hip reconstruction. Methods: Single-surgeon, retrospective analysis of 154 CwCP (204 hips) undergoing hip reconstruction from 2011 to 2024, stratified by those who underwent IHA (IHA: 2016 to 2024) during the index surgery versus those who did not (NIHA: 2011 to 2016). IHA indicated the need for a PO after VDRO, based on labrum orientation and medial dye pooling. Characteristics were compared using χ 2 tests, t tests, and Mann-Whitney U tests, as appropriate. Radiographic outcomes and failure rates (hip resubluxation) >2 years after the index procedure were compared between cohorts, leveraging generalized estimating equations (GEE) modeling and propensity-score-matched (PSM) analyses, adjusting for confounding characteristics. Results: IHA reduced the rate of PO from 82% to 48%, for a 34% reduction, and this was clinically significant ( P =0.001). All hips with upsloping labrums underwent PO, 92% of hips with medial dye-pool underwent PO, whereas downsloping labrums were protective of PO (88.5% underwent VDRO alone). Ten hips failed at a median of 6 years (range: 2.0 to 9.0); however, there were no differences based on ±IHA (GEE-adjusted likelihood of failure 6.3% IHA vs. 8.7% NIHA, P =0.59) or ±PO (6.9% + PO vs. 8.0% VDRO alone, P =0.81). Results of the PSM cohort were equivalent for PO performance (57% IHA vs. 75% NIHA, P =0.03), failure rates, and radiographic outcomes. Conclusions: For patients with CP undergoing hip reconstruction, evaluation of dynamic stability and labral pathoanatomy using IHA reduced the performance of concomitant PO without adversely affecting outcomes. Failures were unrelated to ±IHA or ±PO, suggesting other factors, such as unmitigated hypertonia and/or spinal deformity, were provocative. Level of Evidence: Level III—retrospective comparative study.