Tessa S. Cook MD, PhD, Presenter: Nothing to Disclose

Timothy James Couch, Abstract Co-Author: Employee, Radimetrics Inc

Gregory G. Couch MSc, Abstract Co-Author: Owner, Radimetrics

Woojin Kim MD, Abstract Co-Author: Principal, Montage Healthcare Solutions, Inc Shareholder, Montage Healthcare Solutions, Inc Consultant, Amirsys, Inc Consultant, Infiniti Medical, LLC Speaker, Terumo Corp Principal, iVirtuoso, Inc Shareholder, iVirtuoso, Inc

William W. Boonn MD, Abstract Co-Author: Research Grant, Siemens AG Founder, Montage Healthcare Solutions, Inc Stockholder, Montage Healthcare Solutions, Inc Medical Advisory Board, Merge Healthcare Incorporated

Using anatomy-specific k factors to estimate whole-body effective dose from CT dose-length product (DLP) is practical and widely accepted. However, estimates of organ doses using Monte Carlo (MC) simulations have suggested that the k factor  does not provide an accurate representation of patient dose. In this work, we evaluate the difference between MC-based methods and the k factor-based approach to estimating effective dose (ED) for single-phase abdominopelvic CTs.

496 abdominopelvic CTs (219 male, 277 female) acquired in 2010 were analyzed. ED was estimated by multiplying the ICRP-60 k factor of 0.015 by the total DLP. Monte Carlo simulations using the mathematical human MIRD-5 phantom and average scanner model were also performed to estimate individual organ doses. Total study ED was derived from these organ doses. The ED calculations from both methods were compared, as were individual organ dose estimates for the male and female patient groups.

Average EDs in mSv based on k factor were 12.6+/-5.3 overall, 12.4+/-5.2 for the males and 12.6+/-5.3 for the females. Corresponding average EDs based on MC methods were 16.3+/-7.9, 13.4+/-6.7 and 18.5+/-8.0 mSv. Close correspondence of the ED estimates for the male patients was likely due to the similar size of the MIRD-5 male phantom and the phantom used to derive the k factors. The larger MC-based dose estimates for female patients are likely multifactorial. While the gonads contribute 20% of whole-body dose, the uterus is nearly always included in an abdominopelvic CT, while the testes may only be included if the scan extends inferior to the pubic symphysis. Also, the female patients in this study were smaller on average compared to the male phantom used to derive the k factors, thus resulting in underestimation of ED for female patients when using the k factor.

While estimating ED from total DLP by applying the anatomy-specific k factor is convenient, patient gender and size must be taken into account for accurate estimates. In the absence of real-time derivation of organ doses, these results motivate the need for new, modified k factors that are gender-, size- and anatomy-specific for more accurate ED estimation.

Estimating effective dose with k factors and DLP is convenient and practical, however, care must be taken to acknowledge the limitations of this approach in terms of patient size and patient gender.

Cook, T, Couch, T, Couch, G, Kim, W, Boonn, W, Underestimating Dose to Female Patients Using Dose-Length Product and k Factors for Abdominopelvic CT.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11011982.html