Stacy Kellan Goergen MBBS, Presenter: Nothing to Disclose

Dana J. Jackson RT(R), Abstract Co-Author: Nothing to Disclose

John Heggie PhD, Abstract Co-Author: Nothing to Disclose

Daniel Schick MSc, Abstract Co-Author: Nothing to Disclose

Jane Grimm BS, , Abstract Co-Author: Nothing to Disclose

To measure baseline dose for common exam types delivered by 9 CT scanners, conduct a face-to-face small group optimization workshop, and measure the effect on dose, and in particular, its appropriateness for patient girth. To document in detail protocol variations contributing to higher dose.

Nine MDCT scanners (GE VCT XTe, 1 x GE VCT XT, 2 x GE VCT, GE LightSpeed Pro 16, Philips Brilliance 16,  Toshiba Aquilion 16 X 2, and Toshiba Aquilion 1) were audited. Dose length product (DLP) was the indicative metric. Other parameters collected included: scan range,  patient height, kVp, mAs, pitch, noise presets, acquired / reconstructed slice thickness, dose modulation (x-y, z,), bismuth garment use, patient girth. Data was collected for brain, paranasal sinuses, CTA cerebral, high resolution chest , CT pulmonary angiogram (CTPA),  aportal venous abdomen / pelvis, CT urogram (CTU) and lumbar spine. Data collection occurred over 4 months and 1 day workshop conducted between baseline and post intervention. Feedback about reasons for higher dose per protocol, per scanner was provided in writing for each scanner. Didactic training about contributors to dose was delivered by medical radiation physicists and  radiologist and imaging technologist project leads.  

1800 consecutive studies were audited (928 baseline, 882 post optimization training). Causes of higher than median dose included: girth inappropriate kVp choice, inconsistent noise default setting for acquisition slice thickness, overscanning beyond anatomical requirements, failure to use bismuth garments on compatible scanners, failure to use dose modultation when appropriate. Manufacturer determined variation in default phantom for dose calculation complicated between scanner comparisons. Greatest dose decreases occurred for CTPA (915 - 496), CT abdomen / pelvis (857 - 598), CTU (1026 - 411) and were not accompanied by explanatory reduction in median or range of girth. Increased use of bismuth garments occurred post optimization.

The large number of protocol choices that MDCT requires of technologists complicate optimization  efforts. Benchmarking, protocol specific feedback, and training help achieve dose reductions.

Benchmarking and intensive training of technologists and radiologists regarding scanner  specific protocol modifications can produce clinically important dose reductions without specialized software.

Goergen, S, Jackson, D, Heggie, J, Schick, D, Grimm, J, CT Dose Optimization: What Is the Problem?.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034372.html