AbstractsMedical & Health Science

The purpose of this study was to determine whether radiologists could accurately assess osteopenia on chest plain films. Two chest radiologists evaluated lateral chest films from 100 patients (80 female and 20 male), ranging in age from 16 to 86 years, for osteopenia and its associated findings. Intra- and interobserver agreement was determined using weighted kappa statistics, and accuracy was assessed by making comparisons to bone mineral density as measured by the non-invasive gold standard of dual-energy x-ray absorptiometry (DXA). Overall, radiologists were good at identifying signs of late, but not early, disease. Intraobserver consistency was substantial for fish vertebrae (Kw1=0.638; Kw2=0.0.712) with moderate interobserver agreement (Kw=0.45). Similarly for wedged vertebrae, intraobserver consistency was substantial to moderate (Kw1=0.654; Kw2=0.533) with substantial interobserver agreement (Kw=0.622). These radiographic signs correlated with true disease as shown by high specificity values. Therefore, this study indicates that if osteopenia is suspected (i.e., there is a wedge or fish vertebra) or its associated features are seen on a CXR, it is crucial for radiologists to comment on it. The literature suggests that referring physicians do not pay attention to such findings in radiology reports. Radiologists could effect change in clinical treatment by not burying these findings in the report body, but instead putting it in the impression, along with a recommendation that the finding be followed up with DXA. Because effective interventions for women with osteoporosis exist, the results of this study will contribute to a major change in the practice of chest radiology and improve womens health by preventing the devastating disability associated with osteoporosis.