Association between acetabular coverage over femoral head and rate of joint space narrowing in non-arthritic hips

Abstract

Aims This study was performed to investigate the association between the acetabular morphology and the joint space narrowing rate (JSNR) in the non-arthritic hip. Methods We retrospectively reviewed standing whole-leg radiographs of patients who underwent knee arthroplasty from February 2012 to March 2020 at our institute. Patients with a history of hip surgery, Kellgren-Lawrence grade ≥ II hip osteoarthritis, or rheumatoid arthritis were excluded. The hip JSNR was measured, and the normalized JSNR (nJSNR) was calculated by calibrating the joint space width with the size of the femoral head in 395 patients (790 hips) with a mean age of 73.7 years (SD 8.6). The effects of the lateral centre-edge angle (CEA) and acetabular roof obliquity (ARO) in the standing and supine positions were examined using a multivariate regression model. Results The mean JSNR and nJSNR were 0.115 mm/year (SD 0.181) and 2.451 mm/year (SD 3.956), respectively. Multivariate regressions showed that older age was associated with a larger nJSNR (p = 0.010, standardized coefficient (SC) 0.096). The quadratic curve approximation showed that the joint space narrowing was smallest when the CEA was approximately 31.9°. This optimal CEA was the same in the standing and supine positions. Multivariate regressions were separately performed for joints with a CEA of < 31.9° and > 31.9°. When the CEA was < 31.9°, a smaller CEA was associated with a larger nJSNR (p < 0.001, SC 0.282). When the CEA was > 31.9°, a larger CEA was associated with a larger nJSNR (p = 0.012, SC 0.152). The ARO was not associated with the nJSNR. Conclusion Both insufficient coverage and over-coverage of the acetabulum over the femoral head were associated with increased joint space narrowing in hips that were non-arthritic at baseline. The effects of insufficient coverage were stronger than those of overcoverage.