Comparative study between mobile computed radiography and mobile flat-panel radiography for bedside chest radiography: impact of an antiscatter grid on the visibility of selected diagnostically relevant structures.

ABSTRACT

OBJECTIVES: The objective of this study was to evaluate the diagnostic performance of 2 different imaging systems in adult bedside chest radiography and the impact on the visibility of selected diagnostically relevant structures in the images acquired with these systems, with and without an antiscatter grid. MATERIALS AND METHODS: We acquired bedside chest radiographs of 103 intensive-care patients (36 women, 67 men; age range, 17-90 years; mean age, 66.4 years) using 4 acquisition techniques (computed radiography [CR] and digital radiography [DR], with and without grid). Image quality was evaluated independently by 4 radiologists using a 9-point visibility scale. Evaluated were lung parenchyma, soft tissues, thoracic spine, foreign bodies, and the overall image quality. Interobserver agreement and differences between the systems were tested using an interclass correlation (ICC) test. Mean scores were compared using the analysis of variance, followed by the post hoc pairwise testing (the Tukey test) in case of multiple group comparisons and by the Student t test in case of single group comparisons (P < 0.05, significant). RESULTS: The image quality of the structures evaluated in the DR images with a grid was significantly higher than that obtained without a grid (P < 0.001) for all structures. The use of a grid in CR significantly improved the overall image quality, lung parenchyma, and soft tissue delineation (P < 0.001). Foreign body delineation, however, was significantly better in the CR images obtained without a grid (P < 0.001), whereas the 2 systems showed no significant difference regarding thoracic spine delineation (P = 0.554). The scores of the DR images were significantly higher than those of the CR images for all structures. The interobserver agreement was substantial for lung parenchyma (ICC, 0.77), soft tissue (ICC, 0.78), thoracic spine (ICC, 0.80), and the overall image quality (ICC, 0.78) and was almost perfect for foreign bodies (ICC, 0.81). CONCLUSIONS: The use of an antiscatter grid significantly improved the image quality of bedside DR radiographs. A similar effect was seen with CR radiographs but only for lung parenchyma, soft tissue, and the overall image quality. Mobile DR outperformed CR in all structures.