Comparative Evaluation of Super High-Resolution CT Scan and Virtual Bronchoscopy for the Detection of Tracheobronchial Malignancies: Procedures
The mean (± SD) age of all patients was 52 ± 12.2 years (median, 53 years; range, 29 to 88 years). Thirty-two patients (73%) were men (mean age, 52 ± 11.3 years; median age, 54 years; age range, 29 to 73 years), 12 patients (27%) were women (mean age, 52 ± 14.8 years; median age, 51 years; age range, 38 to 88 years). Histologic diagnoses included non-small cell lung cancer (16 patients), small cell lung cancer (3 patients), metastatic renal cell cancer (12 patients), metastatic melanoma (7 patients), metastatic thyroid cancer (2 patients), sarcoma (1 patient), carcinoid (1 patient), teratoma (1 patient), and esophageal cancer (1 patient). review
For each FB, visualization and interpretation of the tracheobronchial tree were achieved under the direction of the attending thoracic surgeons (DSS and DMN) who were blinded to the radiologist’s interpretation of the imaging modalities. FB findings, which were entered into the database, included the presence or absence of obstructive lesions (ie, a bronchial narrowing of > 50%), endoluminal masses (ie, a mass protruding into the lumen with < 50% occlusion), or mucosal lesions (ie, hemorrhage, erythema, or tissue friability).
For each conventional CT scan, 20 to 50 images of the thorax were obtained, according to the standard of care. The technique was 5-10 collimation, HQ mode (ie, helical pitch, 1 to 3, 11.25 mm table motion per rotation, 120 kVp, 280 mA, 0.8 s tube rotation) and nonoverlapping reconstructions with a section interval of 5 to 10 mm in one held breath using a multislice helical CT scanner.
For each SHR-CT scan, 200 to 300 contiguous 1.25-mm images of the thorax were obtained in one or two 17-s breathhold maneuvers (LightSpeed QX/I multislice helical CT scanner; General Electric; Milwaukee, WI). The technique was 1.25 collimation, HS mode (helical pitch, 6, 7.5 mm table motion per rotation, 120 kVp, 100 mA, and 0.8 s tube rotation), nonoverlapping reconstructions with a section interval of 1.25 mm, and an effective z-axis resolution of approximately 1.6 mm. The manufacturer’s “standard” reconstruction algorithm was used. The multiple scan average dose to the scanned volume was 1.58 rad at the surface and 0.78 rad in the center per examination. The calculated radiation dose with this technique was the same or slightly less than that of a conventional thoracic CT scan.