Is Femoral Head Decentration Reversible After Periacetabular Osteotomy for the Treatment of Developmental Dysplasia of the Hip?

BACKGROUND

Femoral head decentration, seen on MRI as a fluid gap between the femoral head and acetabulum, has recently been introduced as a sign of hip instability. Although periacetabular osteotomy (PAO) effectively corrects bony morphology in developmental dysplasia of the hip (DDH), it is unknown whether femoral head decentration is reversible after acetabular reorientation or whether it persists because of factors such as femoral version or intraarticular degeneration.

QUESTIONS/PURPOSES

(1) Does femoral head decentration, femoral head decentration distance, and its radial extension decrease after PAO for treatment of DDH? (2) Do osseous hip morphologies differ between hips with and without persistent femoral head decentration? (3) Does the severity of joint damage on morphologic indirect magnetic resonance (MR) arthrography of the hip and biochemical cartilage degeneration assessed with the delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) index differ between hips with and without persistent femoral head decentration?

METHODS

We performed a secondary retrospective analysis of a prospectively collected cohort of 34 patients (34 hips) undergoing isolated PAO for symptomatic DDH (lateral center-edge angle < 20°). The original prospective study was designed to evaluate changes in cartilage biochemistry after PAO using serial indirect MR arthrography for dGEMRIC. Thirty-seven patients completed the 1-year MRI follow-up, and 34 had complete imaging data sets available for the present analysis. This cohort was suitable for the present secondary analysis because the standardized MRI protocol included high-resolution radial sequences that enabled assessment of femoral head decentration, osseous morphology, and intraarticular degeneration before surgery and at 1-year follow-up, which was considered sufficient to assess persistent postoperative femoral head decentration after healing and adaptation to acetabular reorientation. Exclusion criteria comprised advanced osteoarthritis (Tönnis Grade ≥ II), neuromuscular disorders, joint incongruity, or incomplete MRI sequences. All patients underwent 1.5-Tesla MRI preoperatively and at 1-year follow-up as part of the source prospective study, utilizing intravenous gadolinium contrast for dGEMRIC and isotropic T2-weighted MRI sequence to assess femoral head decentration and joint morphology. Femoral head decentration was defined as a visible fluid layer on radial MRI slices, quantifying its prevalence, distance (in mm), and radial extension (in degrees around the circumference). The median (IQR) age at surgery was 27 years (20 to 30), and 94% (32 of 34) of patients were female. All patients underwent PAO. Hips with preoperative femoral head decentration were stratified into "resolved" (no femoral head decentration at follow-up) and "persistent" (femoral head decentration at follow-up) groups. We compared groups with resolved and persistent femoral head decentration regarding radiographic parameters (lateral center-edge angle, acetabular index, neck-shaft angle, femoral version), morphologic degeneration (cartilage damage, subchondral cysts, labral tears), and biochemical cartilage damage (mean anterior dGEMRIC index).

RESULTS

Preoperatively, femoral head decentration was present in 29 of 34 of hips. One year after PAO, femoral head decentration completely resolved in 9 of 29 hips and persisted in 20 of 29 hips, with a reduction in median femoral head decentration distance from 4 to 3 mm (p < 0.001) and median radial extension from 90° to 53° (p = 0.003). Preoperatively, hips with persistent femoral head decentration exhibited higher median femoral anteversion (22° versus 10°; p = 0.01) and a higher preoperative acetabular index (24° versus 15°; p = 0.01) compared with hips in which femoral head decentration resolved. Hips with persistent femoral head decentration demonstrated a substantially greater burden of intraarticular degeneration, including severe cartilage damage (13 of 20 versus 1 of 9; p = 0.01), subchondral cysts (14 of 20 versus 0 of 9; p < 0.001), and complex and extensive labral tears (18 of 20 and 17 of 20, respectively, in the persistent group versus 2 of 9 and 3 of 9, respectively, in the resolved group; p < 0.001). Biochemical cartilage health was worse in the persistent group, with a lower median dGEMRIC index (436 msec versus 606 msec; p = 0.004).

CONCLUSION

Although PAO reduced the magnitude and extent of femoral head decentration, complete anatomic resolution occurred in only about one-third of patients with complete follow-up, and it may be less frequent among patients lost to follow-up. Persistent femoral head decentration after PAO was present in hips with excessive femoral anteversion and advanced intraarticular degeneration, including severe cartilage loss, subchondral cysts, and labral lesion. These findings suggest that acetabular reorientation alone may be insufficient to fully restore joint containment in patients with these specific morphologic and degenerative factors. Future studies should further investigate the relationship between femoral head decentration and femoral version abnormalities, which may help define a potential role of concomitant femoral derotational osteotomies.

LEVEL OF EVIDENCE

Level III, therapeutic study.