CT scanning with an adaptive statistical iterative reconstruction (ASIR) algorithm can reduce radiation dose associated with pediatric abdominal scans by an average of 33%, according to a study published online May 19 in Pediatric Radiology.
Pediatric radiologists at the Children's Hospital of Pittsburgh at the University of Pittsburgh Medical Center conducted a retrospective study to evaluate differences in radiation dose and image quality of enhanced abdominal CT scans using ASIR as compared with filtered back-projection (FBP) reconstruction for enhanced abdominal imaging. Images were acquired using either standard routine pediatric dose settings reconstructed with filtered back projection or an additional 38% mA reconstructed with 40% ASIR.
Lead author Dr. Gregory Vorona and colleagues identified 11 patients who had both types of CT scans performed between July 2009 and October 2010, each acquiring images with the same slice thicknesses. The average time between the two studies was 99 days, and the average percentage change in body weight of a patient was 2.4%.
As long as a patient had not grown significantly, images from ASIR and filtered back-projection studies performed at different times could be compared, the authors explained. This allowed for a direct comparison to evaluate radiation dose and image quality differences in the same patients without exposing them to radiation from an additional, unnecessary procedure.
The radiology department had equipped its 64-detector-row CT scanner (LightSpeed VCT, GE Healthcare) with ASIR in July 2009, and had been using 40% ASIR for abdominal studies. At 40% ASIR, dose parameters of the scan were lowered by reducing the manual mA factor to 0.62, resulting in an approximate 38% reduction in mA compared with the department's routine low-dose pediatric imaging protocols.
The CT scanner in the emergency department continued to use filtered back-projection reconstruction. Contrast media injection protocols and the timing of oral contrast were identical regardless of which CT scanner was used.
For the analysis, the authors matched scan parameters such as slice thickness, peak kVp, rotation speed, and pitch. Tube current modulation was not used for any of the scans evaluated in the analysis. For each scan, volume CT dose index (CDTIvol) and dose-length products (DLP) were recorded.
The 40% ASIR studies had an average CDTIvol of 4.25 mGy and a DLP of 185.04 mGy-cm. This compared with a CDTIvol of 6.25 mGy and a DLP of 275.79 mGy-cm for the filtered back-projection studies.
The radiologists' assessments of subjective image quality were comparable for both types of reconstructed images of diagnostic image quality.
