Positioning Essential for CT Dose and Image Noise

A patient being positioned too low during a chest computed tomography (CT) examination could negatively affect the radiation dose needed to image the patient and the image noise, according to the results of a new study. In fact, when using a posteroanterior scout for adults, the lowest position on the table could increase the dose by almost 40 percent.

“We conclude that vertically off-centering of a patient for chest CT is a common and serious problem regardless of patient size,” Touka Kaasalainen, MD, of Helsinki University Central Hospital, Helsinki, Finland, and colleagues wrote in the American Journal of Roentgenology. “Centering the patient vertically below the isocenter of the CT scanner and using a posteroanterior scout for automatic exposure control may significantly increase the dose.”[[{"type":"media","view_mode":"media_crop","fid":"25669","attributes":{"alt":"","class":"media-image media-image-right","id":"media_crop_4980970646720","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"2387","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"line-height: 1.538em; height: 150px; width: 144px; border-width: 0px; border-style: solid; margin: 1px; float: right;","title":" ","typeof":"foaf:Image"}}]]

In the study, Kaasalainen and colleagues sought to assess the effects of vertical off-centering on radiation dose and image quality in chest CT. To do that, they used three anthropomorphic phantoms of different sizes using different vertical centering and commercial radiation dose-monitoring software. They used the phantoms and software to compare the dose of the phantoms to that of 112 patients who had undergone chest CT between November 2012 and March 2013.

In the phantom analysis, when a posteroanterior scout was used for automatic exposure control, the results showed that radiation dose was highest when phantoms were in the lowest position and were lowest when the phantoms were in the highest table position. This discrepancy was lower when lateral scouts were used instead of posteroanterior scouts.

The researchers used CTDI_vol to measure changes in radiation dose. Using the adult-sized phantom, there was a 38 percent increase in relative CTDI_vol when the phantom was positioned 6 cm below the isocenter; a decrease of 23 percent was seen when positioned 6 cm above the isocenter. These same changes were seen in the pediatric and newborn phantoms. A 6 cm shift below the isocenter conferred a 21 percent and 12 percent increased dose, respectively, in pediatrics and newborn phantoms; a 6 cm shift above the isocenter resulted in a 12 percent and 7 percent decreased dose.

“The increases in radiation doses resulting from low off-centering and use of posteroanterior scouts for tube current modulation occurred because the projected area of an object is magnified in the scout image, resulting in a higher tube current-time product for helical scanning,” the researchers wrote. “Moreover, the low off-centering also increases the doses of radiation absorbed by sensitive surface tissues such as the breast and thyroid gland, tissues that are then projected onto the thinnest and least attenuating part of the bowtie filter during the scan rotation.”

When comparing the doses seen in the phantoms to that of the clinical patients, the researchers observed vertical miscentering in all three groups (adult, pediatric and newborn). Observed median values for vertical miscentering varied from 25 to 35 mm below the isocenter. 

The researchers also found that patient centering affected image noise. A low table position increased image noise in the posterior part of the phantoms and decreased noise in the anterior part. In contrast, a high position increased image noise in the anterior and decreased it in the posterior part.

“Special attention should focus on correctly centering patients when preparing patients for CT examinations, especially when imaging children,” the researchers concluded.