Simulated low-dose computed tomography in oncological patients: a feasibility study

Image quality of chest and abdominal computed tomographic (CT) scans was evaluated at different doses to assess the lowest value of x-ray dose at which the image quality was not being affected.

Using multislice CT (MSCT), 29 patients who submitted to follow-up examinations were examined using a combined MSCT protocol of the chest and the abdomen on 4-raw MSCT (Siemens, Erlangen, Germany). For each examination, approximately 120 mL of contrast agent (Bracco-Altana, Konstanz, Germany) was applied intravenously. The raw data were transferred to an external personal computer equipped with an image reconstruction software (CardioRecon 6; Siemens, Forchheim, Germany) to simulate 5 different dose levels. To simulate them correctly, a milliampere second-dependent noise was added to every image, so that the changes in a current-time product could be imitated. The images were compared in consensus by 2 radiologists who were not informed about the technical scanning parameters, that is, dose parameters, and were graded in 4 different subcategories on a 1-to-5-point scale. For statistical analysis, the Friedman test was used. Additional evaluations for lesions smaller than 1.0 cm were performed and analyzed separately.

For image noise, there was a significant change between 40 and 60 mA s. For lesion detection, there was no significant change. The contour of the small objects did not differ between 90, 60, and 40 mA s. However, a dose reduction to 30 mA s had a significant effect. The contrast did not differ between 90, 60, 40, and even down to 30 mA s. Only a maximal dose reduction to 20 mA s had a significant effect on the contrast. The level of noise was most sensitive to the current. Whereas a dose reduction to 60 mA s did not yet have a significant effect, there was a significant increase of noise at 40 mA s.

The MSCT can be applied with a lower dose than that usually selected in examinations at present to follow-up and stage the oncological patients adequately.