Background: Ezoe et al showed that the prevalence of acetabular retroversion in adults with Legg-Calvé-Perthes disease (LCPD) was 42% versus 6% in normal controls. Our purpose was to study the development of acetabular retroversion in children with LCPD and perhaps draw conclusions regarding cause and effect.
Methods: We reviewed all 271 patients with LCPD seen at our institution to identify those patients with axial imaging before closure of the triradiate cartilage. Fifty-three hips (44 patients) formed our study population; the 35 normal hips in those patients with unilateral disease served as an internal control. Acetabular version was measured on the axial cut with the largest femoral head cross-section. Diseased versus control hips were compared using a paired t test. Two disease-severity subgroups, Herring A or B (42 hips) and B/C or C (11 hips), were compared using an independent-samples t test. In those patients followed past skeletal maturity (16 of 53 hips), we evaluated final AP pelvic radiographs for acetabular retroversion (presence of a crossover sign).
Results: Before skeletal maturity, all but 1 patient in our series demonstrated positive acetabular version (anteversion). We found no significant difference between the diseased and control hips: mean acetabular version in LCPD hips was 13.6 +/- 4.3 versus 15.4 +/- 5.4 degrees in unaffected hips. More severe cases (B/C or C) exhibited significantly more relative retroversion (10.8 degrees) than less severe cases (14.6 degrees, P = 0.047). Of the 16 hips followed past skeletal maturity, 5 demonstrated crossover signs on anteroposterior pelvic radiographs, indicating a 31% prevalence of acetabular retroversion.
Conclusion: Early after diagnosis, the prevalence of acetabular retroversion in our skeletally immature children with LCPD was extremely rare (1.8%), and there was no significant difference in version between the normal hips and hips with LCPD. However, over time, a child with a more deformed femoral head is more likely to develop acetabular retroversion, suggesting a cause-and-effect relationship.
Level of evidence: Level III (case-control).