RESUMO
Shelf acetabuloplasty continues to be effective in the prevention or delay of osteoarthritis in adolescent dysplastic hips. We aimed to evaluate the efficacy and to determine the correct level of the bone graft objectively. We retrospectively analyzed 16 hips underwent Shelf acetabuloplasty originated by Spitzy. The mean age at surgery was 15.3 years old with the mean caput index at 39.7%. The average follow-up was 11 years. The radiological evaluation was based on acetabular-head index (AHI) at preoperative and latest radiographs, acetabular-Shelf distance (ASD) and Shelf-head ratio at immediate postoperative and latest radiographs. The cutoff value for the ideal ASD was determined by receiver operating characteristic (ROC) and the Pearson correlation test used in statistical analysis to assess the relationship between ASD and Shelf graft resorption. Clinical evaluation was performed using Harris Hip Score (HHS) at the latest follow-up. The AHI was improved in all cases, from mean 56.9 to 91.0% (P < 0.001). The mean of ASD was 7 mm. In nine of the 16 cases, the Shelf graft was well united at the same level of the existing acetabulum with good continuity. The ROC curve showed the cutoff value for the ideal ASD was 6 mm. The Pearson correlation test also showed a positive relationship between ASD and Shelf graft resorption (P = 0.001). The average of HHS scores was 98.1 points. Both radiological and clinical results were acceptable. The ideal placement that keeps the shelf sufficient to bear the mechanical stress without bone resorption over time was at the level of 6 mm from the joint space. Level III - therapeutic study.