PET/CT: 32% dose reduction possible with CT optimization

Standardized CT protocols can lead to a substantial decline in radiation dose, according to a large-scale study published in the August issue of American Journal of Roentgenology.

A team of researchers including Elena Tonkopi, a diagnostic imaging physicist at Dalhousie University and the Queen Elizabeth II Health Sciences Centre in Halifax, Nova Scotia, Canada, reviewed whole-body PET/CT scans of 240 patients to determine how much reduction in radiation dose was possible with CT optimization.

“Various methods and strategies based on CT technology have been explored for dose reduction,” wrote Tonkopi et al. “When CT is used only for attenuation correction of the PET data, the range of exposure settings that can be used is extremely wide; however there are limitations if the CT data are also used for anatomic localization. Careful consideration of the acquisition parameters is required to achieve low patient dose with acceptable image quality.”

The average CT-related dose to the pelvis is about 8 mSv followed by 7 mSv for the chest and 3 mSv for the head. A whole-body scan might be about 24 mSv of CT-related dose and a total of 46 mSv when the PET radiopharmaceutical is included. For this study, researchers reviewed 140 consecutive patient scans and then compared the radiation dose to a second cohort of 100 patients, with some crossover, imaged under a strict CT protocol. Optimization included shorter scan times and lower tube current. Researchers increased the noise index from 25 to 27.1, slightly lowered pitch and rotation time to 1.35 and .05, respectively, and adjusted beam collimation to 16 x 1.25.

“This dose reduction was achieved with a faster x-ray tube rotation time, increased x-ray beam coverage, and higher noise index value,” wrote the authors. “With the chosen beam collimation, the pitch value had to be decreased because only certain combinations of detector configurations and pitches are allowed on the scanner model we use. However, the overall effect of optimization resulted in decreased radiation dose. Implementation of this new protocol is justified by the results of our image quality evaluation.”

A total of 26 patients from the original two groups were chosen for a blind before-and-after image analysis, which resulted in no significant decline in imaging quality. The total reduction of CT dose was from 8.1 to 5.5 mSv. The only areas that were found to be optimal at a higher dose were the carotid arteries and the posterior triangle of the neck.

The authors cautioned that this protocol was specific to one model of PET/CT and that institutions may have to tweak their own protocols for best results.