DIAGNOSTIC REFERENCE LEVELS IN DIGITAL MAMMOGRAPHY: A SYSTEMATIC REVIEW Moayyad E. Suleiman*, Patrick C. Brennan and Mark F. McEntee Faculty of Health Sciences, The University of Sydney, M205, Cumberland Campus, 75 East St, Lidcombe, Sydney, NSW 2141, Australia *Corresponding author: msul1801@uni.sydney.edu.au Received 7 October 2014; revised 24 November 2014; accepted 3 December 2014 This studyaims to review the literature on existing diagnostic reference levels (DRLs) in digital mammographyand methodolo- gies for establishing them. To this end, a systematic search through Medline, Cinahl, Web of Science, Scopus and Google scholar was conducted using search terms extracted from three terms: DRLs, digital mammography and breast screen. The search resulted in 1539 articles of which 22 were included aftera screening process. Relevant data from the included studies were summarised and analysed. Differences were found in the methods utilised to establish DRLs including test subjectstypes, proto- cols followed, conversion factors employed, breast compressed thicknesses and percentile values adopted. These differences com- plicate comparison of DRLs among countries; hence, an internationally accepted protocol would be valuable so that international comparisons can be made. INTRODUCTION Breast cancer causes almost half a million deaths in the world per year (1) , but early detection has been demonstrated to reduce mortality by up to 30 % (2) . Mammography, radiographic imaging of the breast with X rays, is the most important diagnostic tool for the early detection of breast cancer. There are two types of patients on whom mammograms are per- formed: symptomatic women in the clinic and asymp- tomatic women in breast screening programmes. The Australian breast screening programme was established in 1991, targeting women aged 50–69 y for 2-yearly screening mammograms with the aim of redu- cing deaths from breast cancer (3) . It has been estimated that since 1991 breast cancer mortality in Australia has been reduced by 21–28 % (3) ; however, as with any other X-ray examination, screening programmes can add to the risk of inducing cancer in healthy women by exposure to ionising radiation. Therefore, the dose to the patient must be kept as low as reasonably achiev- able (4) . The three pillars of radiation protection are justification, optimisation and dose limitation. The International Commission of Radiation Protection (ICRP) introduced diagnostic reference levels (DRLs) in their 1996 publication 73 as a par- ameter to be used for quality control, comparison of dose levels, optimisation and limiting variations in dose among diagnostic imaging centres. DRLs were defined as follows: A form of investigation level, applied to an easily measured quantity, usually the absorbed dose in air, or tissue-equivalent material at the surface of a simple phantom or a representative patient (4) . A year later, the European Council defined DRLs as: Dose levels in medical radiodiagnostic practices, for typical examinations for groups of standard- sized patients or standard phantoms for broadly defined types of equipment. These levels are expected not to be exceeded for standard proce- dures when good and normal practice regarding diagnostic and technical performance is applied (5) . The methods through which the DRLs are established become important when trying to establish inter- national comparisons as radiation dose measure- ments are required (3) . Historically, mammography was screen-film based (6 – 12) , but now this technology is being phased out and replaced with digital mam- mography, which includes full-field digital mammog- raphy and computerised radiography systems; hence, only studies with digital mammography or a mix of digital mammography and screen-film mammography (SFM) are included. Measuring the radiation dose to the breast has been performed or represented using a variety of approaches including air kerma (13) , entrance surface dose (14) , mid-breast dose (15) , total energy trans- mitted to the breast (16) and the average dose absorbed by the glandular tissue (17) . The latter was found to be the most effective way of measuring absorbed dose to the breast because the mammary glands are most sensi- tive to ionising radiation and have the highest risk of developing radiation-induced carcinogenesis (17) . Called mean glandular dose (MGD), this metric is now the recommended measure by many authorities such as the ICRP (18) , the United States National Council on Radiation Protection and Measurements (19) , the British Institute of Physics and Engineering in Medicine # The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com Radiation Protection Dosimetry (2014), pp. 1–12 doi:10.1093/rpd/ncu365 Radiation Protection Dosimetry Advance Access published December 27, 2014 by guest on December 28, 2014 http://rpd.oxfordjournals.org/ Downloaded from