The distribution and inter-relationships of radiologic features of osteoarthrosis of the hip. A survey of 4151 subjects of the Copenhagen City Heart Study: the Osteoarthrosis Substudy.

OBJECTIVE

The aims of this study were to investigate the influence of sex, age and individual physical and occupational factors on the distribution of radiographic features of hip joint osteoarthritis (OA), and to determine the inter-relationships between the primary radiographic OA discriminator of reduced joint space width (JSW), and secondary radiographic OA features.

METHOD

The study was a cross-sectional survey. Radiologic features of hip OA were recorded from standardized, standing pelvic radiographs of 3686 subjects (1397M/2289F). OA features were investigated for inter-relationships and correlations to age, sex, body mass index (BMI) and occupational exposure to repeated lifting.

RESULTS

Overall, subchondral sclerosis, cysts and osteophytes were more frequently recorded in male hip joints compared to female hip joints, while a decrease in minimum JSW by age was more pronounced and progressive in women after the fifth decade compared to men. Applying logistic regression analyses, only age was found to be significantly associated to pathologically reduced minimum JSW (cut off value set at < or =2.0 mm), and the presence of osteophytes and subchondral cysts in both sexes (P ranging from 0.00 to 0.03). Minimum JSW < or =2.0 mm was recorded in 105 male hip joints, and in 167 female hip joints. In these joints, subchondral cysts, osteophytes and sclerosis were found to be significantly inter-related to minimum JSW < or =2.0 mm according to logistic regression analysis. The presence of subchondral cysts had the highest predictive sensitivity in regard to pathologically reduced minimum JSW compared to subchondral sclerosis and osteophytes.

CONCLUSIONS

We believe that an accurate radiologic case definition of hip OA will rest on a combination of features. Only studies of the distribution of features, singularly and in combination, applied to clinically evaluated large cohorts may provide the optimum answer to the best possible clinical case definition of hip OA.