Is there any diagnostic value of anteroposterior chest radiography in predicting cardiac chamber enlargement?

Abstract

The anteroposterior (AP) portable chest radiograph is routinely performed to evaluate cardiopulmonary status, however heart size can be misrepresented by inherent technical factors. Our aim was to determine diagnostic accuracy of cardiothoracic ratio (CTR) on AP chest radiographs relative to echocardiography, as well as relative to axial computed tomography (CT) and frontal CT scout images in predicting cardiac chamber enlargement. 200 subjects with both chest CT and AP chest radiograph within 1month were retrospectively identified. Patients with pericardial effusion or obscured heart borders were excluded. 130 of these subjects had also undergone echocardiography. Transverse diameters of the heart and thorax were used to calculate CTRs on AP chest radiograph, scout CT, and axial CT images. A second reader was used to verify measurement accuracy and reproducibility. Statistical analysis of CTRs for AP chest radiograph, CT scout, and axial CT images were calculated using echocardiography as gold standard. AP chest radiographs had higher CTR values than axial and scout CT images (by 0.075, p < 0.001), larger measured heart diameters by approximately 3cm (p < 0.001), and larger thoracic diameters by approximately 2cm (p < 0.001). CTRs on AP chest radiographs calculated with a cutoff of 0.50 had sensitivity of 86% and specificity of 32%. Sensitivity and specificity were 61% and 66% respectively when using a cutoff of 0.55, and 34% and 92% respectively when using a cutoff of 0.60. A CTR of 60% is more appropriate than 50-55% when evaluating an AP chest radiograph for cardiac chamber enlargement due to its much higher specificity.