Utilizing Graphical Analysis of Chest Radiographs for Primary Screening of Osteoporosis

Abstract

Background and Objectives: Osteoporosis is a major risk of fractures, harming patients’ quality of life. Dual-energy X-ray absorptiometry (DXA), which can detect osteoporosis early, is too expensive to be conducted on a regular basis. Therefore, we aimed to evaluate a screening method using chest radiographs developed in Japan applied to another population. Materials and Methods: Fifty-five patients who had a chest radiograph and DXA and applied within three months of each test were recruited from the patient database of Semmelweis University (Budapest, Hungary). Graphical analysis of the chest radiographs was conducted to identify the ratio of the cortical bone in the clavicle of each patient. Two researchers performed the analysis, and multiple regression was conducted to determine the bone mineral density of each patient provided by DXA. Results: The Pearson correlation between two examiners’ determinations of the cortical bone ratio was 0.769 (p < 0.001). The multiple regression model proved to be statistically significant in identifying osteoporosis, but the model adopted for the Hungarian population was different compared to the Japanese population. Conclusions: This simple, economic Japanese graphical analysis method for chest radiographs may be feasible in detecting osteoporosis. Further studies with a larger population of patients with greater variety of ethnicity would be of value in improving the accuracy of this model.