A Possible New Radiographic Predictor of Progression of Osteoarthritis in Developmental Dysplasia of the Hip: The Center Gap.

BACKGROUND: Patients with comparable severities of developmental dysplasia of the hip (DDH) may variably progress to osteoarthritis (OA) over time. Although joint congruency may be associated with OA progression in patients with DDH, it has only been assessed subjectively. We assessed the gap between the rotational centers of the acetabulum and femoral head (center gap) as a possible predictive measure of OA progression in patients with DDH. QUESTIONS/PURPOSES: In patients with bilateral DDH, we asked: (1) What is the probability of OA progression (Tönnis grade) or symptom development (pain) in the asymptomatic contralateral hip of patients with DDH undergoing unilateral joint-preserving surgery? (2) Is the center gap measurement associated with OA progression or symptom development in these hips? (3) Is the center gap measurement correlated with previous radiographic parameters?
METHODS: A total of 297 patients (319 hips) underwent unilateral eccentric rotational acetabular osteotomy at our institution between July 1989 and December 1999. We performed no other joint-preserving surgery to treat patients with DDH during this timeframe. The inclusion criteria for the study were patients younger than 55 years of age, the contralateral hip classified as Tönnis Grade 0, no previous surgical interventions, and asymptomatic at the time of eccentric rotational acetabular osteotomy (155 patients, 155 hips). The exclusion criteria were a contralateral hip without dysplasia (four patients, four hips), loss to followup before 10 years (42 patients, 42 hips [27%]), or missing medical records or radiographs (21 patients, 21 hips [14%]). The remaining 88 patients (88 hips; 11 males and 77 females) with a mean age of 39 years (range, 17-53 years) and mean followup of 20 years (range, 10-27 years) were analyzed. From the institutional database, radiographic parameters including the center gap in the AP view were assessed using radiographs at the time of eccentric rotational acetabular osteotomy, and the Tönnis grade was recorded 1 year postoperatively and annually thereafter retrospectively. We defined migration of the rotational center of the femoral head based on the rotational center of the acetabulum in the horizontal plane as center gap X (mm) and in the vertical plane as center gap Y (mm) and defined the absolute value between the centers as center gap distance (mm). Using κ statistics, intra- and interobserver reliabilities were determined to be 0.896 and 0.857 for center gap X, 0.912 and 0.874 for center gap Y, and 0.912 and 0.901 for the center gap distance, respectively. When patients reported any contralateral ipsilateral hip pain during clinic visits, the hip was considered symptomatic. Kaplan-Meier survivorship analyses were performed with OA progression or symptom development in the nonoperative hip as the endpoint. Multivariate analyses were performed to assess risk factors for each outcome using the Cox proportional hazards model. Correlation analyses between the center gap and other parameters including lateral center-edge angle, femoral head extrusion index, acetabular depth-to-width index, femoral head lateralization, minimum width of the joint space, head sphericity, and joint congruency were performed using Pearson's correlation coefficient.
RESULTS: At 20 years postoperatively, the probability of OA progression in the nonoperative hip was 13% (95% confidence interval [CI], 7.1-22.1) and the probability of symptom development was 34% (95% CI, 24.7-46.1). The center gap X measurements in the groups with OA progression (lateral 2.0 ± 2.1 [SD] mm) or symptom development (lateral 0.9 ± 2.4 mm) took a more lateral direction than those in the group without OA progression (medial 0.4 ± 2.1 mm) or symptom development (medial 0.5 ± 2.0 mm) (OA progression, p < 0.001; symptom development, p = 0.005). The center gap Y measurements in the groups with OA progression (distal 2.7 ± 7.1 mm) or symptom development (distal 2.1 ± 6.0 mm) took a more distal direction than those in the group without OA progression (proximal 1.6 ± 6.2 mm) or symptom development (proximal 2.5 ± 6.1 mm) (OA progression, p = 0.027; symptom development, p = 0.001). Independent risk factors for OA progression were the femoral head extrusion index (hazard ratio [HR], 1.11; 95% CI, 1.01-1.22; p = 0.028) and the center gap X (HR, 1.52; 95% CI, 1.07-2.16; p = 0.019), whereas no independent risk factors for symptom development were found. The center gap in the horizontal plane had no correlations with any other radiographic parameter studied.
CONCLUSIONS: The center gap in the horizontal plane had a modest association with OA progression in this group of patients with DDH. Future studies are needed to determine the normal value of the center gap for patients without DDH and to assess the center gap in lateral radiographic views. LEVEL OF EVIDENCE: Level IV, prognostic study.