Marilyn J. Goske MD, Presenter: Nothing to Disclose

Keith J. Strauss MS, Abstract Co-Author: Nothing to Disclose

Thomas Louis Toth, Abstract Co-Author: Consultant, General Electric Company Former employee, General Electric Company Co-owner, Plexar Associates, Inc Employee, Plexar Associates, Inc

Michael F. McNitt-Gray PhD, Abstract Co-Author: Institutional research agreement, Siemens AG Research grant, Siemens AG Instructor, Medical Technology Management Institute

Cynthia H. McCollough PhD, Abstract Co-Author: Research grant, Siemens AG

John M. Boone PhD, Abstract Co-Author: Consultant, Varian Medical Systems, Inc Consultant, Artemis, Inc Research funded, Varian Medical Systems, Inc Research funded, FUJIFILM Holdings Corporation Research funded, Hologic, Inc Research funded, Creatv MicroTech, Inc Research funded, Siemens AG

Dianna D. Cody PhD, Abstract Co-Author: Nothing to Disclose

Donald P. Frush MD, Abstract Co-Author: Nothing to Disclose

Estimating the radiation dose a child receives before or after a CT scan is challenging due to a wide range in body size. Scanner radiation output displayed as Volume CT Dose Index (CTDIvol) can be a factor of 3 lower than the actual dose to an infant. Thus, CTDIvol is a poor surrogate for patient dose in pediatric CT. We describe a Size Specific Dose Estimate (SSDE) that allows users to estimate pediatric patient doses based on body size.  

Four independent medical physics teams used different methods to derive conversion factors that when applied to conventional CTDIvol values (16 or 32 cm PMMA phantoms) improved accuracy of radiation dose estimation. The four methods were 1) physical measurements of anthropomorphic phantoms, 2) physical measurements using cylindrical PMMA phantoms, 3) Monte Carlo simulations of voxelized phantoms and 4) Monte Carlo measurements on simple cylindrical phantoms. Data were obtained with different tube potentials and scanner models. The data were combined and fit to an exponential relationship (Y= a * e^bx) to determine if a single relationship could be used to convert CTDIvolto size specific dose.  

Combined data demonstrated remarkable agreement and were highly correlated to an exponential function (R^2 = 0.942). This suggests that only one set of size-dependent conversion factors is necessary across tube voltage and scanner manufacturer to accurately estimate patient dose from CTDIvol values. Conversion factors generated from the curve are provided in look up tables from American Association of Physicists in Medicine Taskgroup (TG) 204. Requirements for the user to generate SSDE (accurate to 10-20%) is knowledge of a patient body dimension (AP or lateral thickness or effective diameter) measured from the scout image (scanned projection radiograph) or axial image, the phantom size used to generate CTDIvol, and CTDIvol. Instructions for use and suggested phrasing for documenting SSDE in the radiology report are provided.  

TG 204 developed easy-to-use look up tables that can be applied to the displayed CTDIvol which allows radiologists to easily estimate patient dose for children and small adults.

Past estimates of pediatric patient dose have been cumbersome to generate or inaccurate. SSDE is a major advance for patient care and has relevance for dose registries and documentation.

Goske, M, Strauss, K, Toth, T, McNitt-Gray, M, McCollough, C, Boone, J, Cody, D, Frush, D, Size-specific Dose Estimate (SSDE): A More Accurate Method of Estimating Patient Dose from CT for Children and Small Adults.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11007269.html