Chest wall infiltration by lung cancer: value of thin-sectional CT with different reconstruction algorithms

Abstract.

The aim of this investigation was to evaluate whether thin-sectional CT with different reconstruction algorithms can improve the diagnostic accuracy with regard to chest wall invasion in patients with peripheral bronchogenic carcinoma. Forty-one patients with intrapulmonary lesions and tumor contact to the thoracic wall as seen on CT staging underwent additional 1-mm CT slices with reconstruction in a high-resolution (HR) and an edge blurring, soft detail (SD) algorithm. Five criteria were applied and validated by histological findings. Using the criteria of the intact fat layer, HRCT had a sensitivity of 81 % and a specificity of 79 %, SD CT had a sensitivity of 96 % and a specificity of 78 %, and standard CT technique had a sensitivity of 50 % and a specificity of 71 %, respectively. Regarding changes of intercostal soft tissue, HRCT achieved a sensitivity of 71 % and a specificity of 96 %, SD CT had a sensitivity of 94 % and a specificity of 96 % (standard CT technique: sensitivity 50 % and specificity 96 %) . For the other criteria, such as pleural contact area, angle, and osseous destruction, no significant differences were found. Diagnostic accuracy of chest wall infiltration can be improved by using thin sectional CT. Especially the application of an edge-blurring (SD) algorithm increases sensitivity and specificity without additional costs.