| dc.description.abstract | Introduction: Mammography is the most effective screening modality for breast cancer due to the early detection of breast lesions. Breast tissue is highly sensitive to radiation. The risk of radiation-induced breast cancer following mammography increases with the average glandular dose (AGD). In 1996, the International Commission on Radiological Protection (ICRP) introduced the Diagnostic Reference Level (DRL) to optimize the dose for all ionizing radiation including mammography.
Objectives: To establish the National Diagnostic Reference levels (NDRLs) for digital screening and diagnostic mammography in Uganda.
Methods: A cross-sectional study design was conducted in three hospitals offering digital mammography services in Uganda with a total sample size of 300 participants. The study variables extracted from the Digital Imaging and Communications in Medicine (DICOM) header were kilovoltage peak (kVp), milliampere-seconds (mAs), compressed breast thickness (CBT), breast compression force (BCF), Entrance Surface Dose (ESD) and Average Glandular Dose (AGD) of two standard mammogram views for each breast. The national DRL was calculated using the 75th percentile median value of the Averaged Glandular Dose (AGD) for three CBT ranges across all three facilities.
Results: The mean ages of the patients for the screening and diagnostic mammography were 50.28±9.32 and 47.45±13.45 years respectively.
There was a significant positive correlation between AGD and kVp, CBT in both screening and diagnostic mammography of each view. There was a strong positive correlation between AGD and mAs in both craniocaudal (CC) and mediolateral oblique (MLO) views for screening mammography (r =0.8369 and 0.8133, and p-values of <0.0001 and <0.0001 in CC and MLO respectively) and diagnostic mammography (r=0.8987 and 0.8762, p-values of <0.0001 and <0.0001 in CC and MLO projections respectively). However, there was a negative correlation between AGD and BCF in the CC view in the screening mammography (r = -0.1993, p-value= 0.0145) and no correlation with the MLO view. There was no correlation between BCF and AGD in both mammographic views in the diagnostic mammography modality (r=0.0598 and 0.0220, p-values= 0.4676 and 0.7894 in CC MLO projections respectively).
The national DRLs for the digital screening and diagnostic mammography were established at a stratified CBT in CC and MLO views as follows; for screening mammography at CBT of (7-39) mm, DRL = 1.5 mGy and 1.66 mGy, CBT of (40-59) mm, DRL=1.78 mGy and 1.87 mGy and at CBT of (60-99)mm DRL= 2.18 mGy and 2.22 mGy for CC and MLO views respectively. While for diagnostic mammography at CBT of (7-39) mm DRL= 1.7mGy and 1.91mGy, at CBT of (40-59) mm, DRL 2.00mGy and 2.09mGy, and at (60-99) mm, DRL =2.63mGy and 2.81mGy respectively in CC and MLO views respectively. The national DRL values for diagnostic mammography obtained in this study were well higher than those for screening mammography for the corresponding CBT and projections.
Conclusion
The National DRL values in mammography should be specific to compressed breast thickness and projections for both diagnostic and screening mammography. The optimization of the exposure parameters according to the compressed thickness is essential to optimize the AGD | en_US |
