Farzana Mehzabin Rahman FRCR, MBBS, Presenter: Nothing to Disclose

Raj Das MBBS, FRCR, Abstract Co-Author: Nothing to Disclose

Linda Howarth BSC, Abstract Co-Author: Nothing to Disclose

Jerome Valette, Abstract Co-Author: Employee, General Electric Company

Ioannis Vlahos MRCP, FRCR, Abstract Co-Author: Consultant, Siemens AG Consultant, General Electric Company

As MDCT protocols continuously evolve, significant variations in patient radiation exposure due to different scanners and/or protocols may take time to appreciate, particularly if monitoring is limited to comparing sample “standard sized” patients to dose reference ranges. We aim to evaluate the impact of a software auditing tool that automatically compiles the total radiation dose of all patient CT examinations performed (Dosewatch, GE Healthcare).

Dosewatch was implemented for 10 weeks, monitoring the dosimetry of all studies performed on two identical 64-detector MDCT scanners (CT1, CT2, Lightspeed VCT, GE Healthcare) believed to be running identical CT protocol iterations. Both scanners were national reference standard compliant. CT Dose Length Product (DLP) was compared between both scanners across all spiral acquisitions according to protocol and body part with large DLP protocol variations explored.

3459 spiral acquisitions (mean DLP 467mGy-cm) using 64 different protocols were recorded. The two commonest CT protocols (Chest/Abdomen/Pelvis, n=489 and Brain ,n=479) had CT2 doses within 3% and -1% of CT1 (826 v 803 and 298 v 301mGy-cm respectively). The 10 most common examinations (2018 spiral acquisitions, 58%) demonstrated only 4% DLP variation. Review identified 3 CT protocols with potential significant CT2 v CT1 DLP discrepancy: Pulmonary Angiography (CTPA) 171 v 213 (-24%), Colonography (CTC) 699 v 587 (-19%) and Renal Arteries (CTRA) 172 v 124 (-38%). Protocol comparison revealed identical coverage, kVp and slice thickness. However, maximum mA setting was inadvertently set higher for CTPA on CT1 (700 v 650mA) and the pitch was inadvertently set lower on CT1 for CTC (0.97 v 1.4). For CTRA (2 spirals/exam), an inadvertently higher set noise index reducing DLP on CT2 (320 v 300) was partially offset by a lower pitch (0.52 v 0.98) on the other spiral.

A CT dosimetry automated global tracking confirmed exam DLP was closely comparable in two identical systems, yet due to the large number of protocols evolved over time, 3 protocols had subtle correctable undetected differences resulting in significant patient dose variation.

The use of global automatic dose monitoring software helps identify variations in protocols even in identical CT systems that may have significant implications for patient exposure.

Rahman, F, Das, R, Howarth, L, Valette, J, Vlahos, I, Impact of a Comprehensive Automated Continuous Dosimetric Monitoring Tool on Departmental Patient MDCT Radiation Exposure and Protocol Optimization.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12033509.html