Average Glandular Dose in FDA-approved FFDM Systems
The Food and Drug Administration’s (FDA) Office of Device Evaluation (ODE) approved the first Full Field Digital Mammography (FFDM) unit for marketing in January 2000. At that time, no Accreditation Body (AB) had developed the necessary standards to accredit any FFDM systems. As a result, FDA’s Division of Mammography Quality and Radiation Programs (DMQRP) could not approve an AB to accredit such units. Consequently, DMQRP developed a temporary procedure to allow for the clinical use of these FFDM units at facilities that were already certified for screen-film mammography (SFM). According to this procedure, an SFM certified facility that acquired an ODE-approved FFDM unit had to apply to DMQRP to extend its SFM certification to include this FFDM unit(s). Such units were referred to as FDA-approved FFDM units. Before DMQRP could approve such use, the facility had to submit an application indicating that each of the mammography personnel providing services using this FFDM unit has met the 8 hours of initial new modality training, and a report of a physics survey or a mammography equipment evaluation (MEE), conducted on the unit by a medical physicist, indicating that the new FFDM unit had passed all the tests required under the final regulations. One of those tests consisted of measuring the average glandular dose for a cranio-caudal view of the standard breast, as simulated by the accreditation phantom, in order to verify that the dose did not exceed 300 mrad.
Between June 2, 2000 and September 2, 2003, DMQRP extended the certificates of 264 screen-film facilities to cover a total of 337 FFDM units according to the procedure described above. The following is a summary of the pertinent data regarding the average glandular dose recorded by medical physicists as part of the physics survey/MEE for 318 of these units.
The FFDM units included in this summary are: the General Electric 2000D, the Fischer SenoScan, and the Lorad Selenia. Since the Automatic Exposure Control (AEC) system in the 2000D has multiple modes for what GE refers to as “Automatic Optimization of Parameters (AOP),” resulting in different dose values, we only used results from the Standard AOP mode for this summary, or the procedure recommended by GE for measuring the dose to the phantom, where applicable. The results are summarized below:
Maximum Dose: 298 mrad
Minimum Dose: 57 mrad
Average Dose: 151.8 mrad
Standard Deviation: 36.5 mrad
Coefficient of Variation: 0.23
The average dose recorded in these FFDM units is about 15% lower than the dose of screen-film units measured by MQSA inspectors during the same time period.
A histogram of the dose data described above is shown in the graph below:
The histogram shows a broad and slightly uneven distribution towards cases with dose values above the average.
Other pertinent information complementing this histogram is the quartile data, which shows the maximum dose value for each quartile (e.g., each group having 25% of all the data). This data is shown below:
First Quartile: 126
Second Quartile: 146
Third Quartile: 172
Fourth Quartile 298