Emilio Quaia MD, Abstract Co-Author: Nothing to Disclose

Elisa Baratella MD, Presenter: Nothing to Disclose

Gabriele Poillucci, Abstract Co-Author: Nothing to Disclose

Sara Kus, Abstract Co-Author: Nothing to Disclose

Antonio Giulio Gennari, Abstract Co-Author: Nothing to Disclose

Maria Assunta Cova MD, Abstract Co-Author: Nothing to Disclose

To assess the actual clinical impact of digital tomosynthesis in the management of oncologic patients with suspected pulmonary lesions on chest radiography.

Two-hundred-thirty-seven patients (135 male, 102 female; age, 70.8±10.4 years) with a known primary malignancy (colorectal, n=49; breast, n=46; bladder, n=22; kidney, n=22; lung, n=20; prostate, n=16; stomach, n=8; non-Hodgkin lymphoma, n=4; others, n=50) suspected pulmonary lesion(s) on chest radiography underwent digital tomosynthesis. Two readers (experience, 10 and 25 years) prospectively analyzed chest radiography and digital tomosynthesis images and proposed a diagnosis according to a confidence score: 1 or 2=definite or probable benign pulmonary or extra-pulmonary lesion, or pulmonary pseudolesion deserving no further diagnostic work-up; 3=indeterminate; 4 or 5=probable or definite pulmonary lesion deserving further diagnostic work-up by CT. DTS findings were proven by imaging follow-up by CT (n=124 patients) or chest radiography (n=106) or histology (n=7). Mean interpretation time and effective dose were measured both for chest radiography and digital tomosynthesis.

Final diagnoses included 94 pulmonary lesions, 14 pulmonary scars and 14 pleural lesions in 122 patients, and pulmonary pseudolesions in the remaining 115 patients. Digital tomosynthesis resolved the chest radiography doubtful findings in 123/237 (52%) patients, while 114/237 (48%) patients underwent CT. Digital tomosynthesis vs chest radiography presented an higher sensitivity (88% vs 15%), specificity (95% vs 10%), overall accuracy (93% vs 21%) and diagnostic confidence (area under ROC curve, 0.914 vs 0.558). The mean interpretation time for digital tomosynthesis (mean±SD, 220 ± 40 s) was higher (P<0.05; Wilcoxon test) than for chest radiography (110 ± 30 s) but lower than CT (600 ± 150 s). Mean effective dose was 0.06 mSv (range, 0.03 – 0.1 mSv) for chest radiography, and 0.2 mSv (range, 0.1 – 0.3 mSv) for digital tomosynthesis.

Digital tomosynthesis avoided the need for chest CT in about one half of oncologic patients with suspected pulmonary lesions on chest radiography with a slight increase in the interpretation time and effective dose comparable to chest radiography.

Digital tomosynthesis is a problem-solving imaging technique to rule out suspicious pulmonary findings in oncologic patients with an high pre-test probability.

Quaia, E, Baratella, E, Poillucci, G, Kus, S, Gennari, A, Cova, M, Diagnostic Impact of Digital Tomosynthesis in Oncologic Patient Management: Analysis on a Large Patient Series.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14003319.html