Digital tomosynthesis as a problem-solving imaging technique to confirm or exclude potential thoracic lesions based on chest X-ray radiography

Abstract

Rationale and objectives: To assess the capability of digital tomosynthesis (DTS) as a problem-solving imaging technique to confirm or exclude potential thoracic lesions based on chest x-ray radiography (CXR).

Materials and methods: Four hundred and-sixty five patients (263 male, 202 female; age, 72.47 ± 11.33 years) with suspected thoracic lesion(s) after the initial onsite analysis of CXR underwent DTS. Two independent readers prospectively analyzed in consensus CXR and DTS images on a picture archiving and communications system-integrated workstation and proposed a diagnosis according to a confidence score for each lesion: 1 or 2 = definite or probable pulmonary or pleural benign lesion or pseudolesion deserving no further diagnostic work-up; 3 = indeterminate; 4 or 5 = probable or definite pulmonary lesion deserving further diagnostic work-up by computed tomography (CT). In patients who did not undergo chest CT, the DTS findings had to be confirmed by 6 to 12 months' imaging follow-up.

Results: Finally, 229 pulmonary lesions (193 thoracic and 36 pleural lesions) and 236 pseudolesions were identified. Based on DTS images, readers correctly classified all pseudolesions except for 10/236 (reader 1) or 11/236 (reader 2) pseudolesions and 7 (reader 1) or 6 (reader 2) pulmonary subpleural lesions located in the anterior or posterior lung region close to the thoracic wall. Chest CT was performed in 127/465 (27%) patients, whereas in 338/465 patients (73%) CXR doubtful findings were resolved by DTS.

Conclusions: DTS allowed to exclude most pseudolesions initially considered as potential thoracic lesions on the preliminary onsite assessment of CXR and allowed to exclude pulmonary lesions deserving CT assessment in about three fourths of the patients.