Author: Emi Ai Eastman, Christina Lee, Xinhua Li, Yifang (Jimmy) Zhou 👨🔬
Affiliation: Cedars-Sinai Medical Center 🌍
Purpose:
This study aimed to retrospectively evaluate dose reduction efforts in past five years using acquisition-level data and to compare the results with ACR achievable dose (AD) and dose reference level (DRL).
Methods:
A holistic approach to protocol review and approval for dose reduction has been employed at our institution. It utilizes a dose tracking software (ImalogixTM) with customized dose thresholds, an online protocol book, and PACS to analyze individual alerted cases. Monthly monitoring of aggregate dose distributions was conducted to identify population trends. Median and 75th percentile CTDIvol and DLP at acquisition level for top 10 adult CT protocols were analyzed retroactively. This study included 271,156 acquisitions from 263,252 exams, originating from 10 CT scanners. Exams were grouped by quarters according to their dates. The results were compared to ACR AD and DRL.
Results:
Over 5 years, eight protocols showed decrease in median and 75th CTDIvol. By Q4 2024, median CTDIvol for nine protocols and median DLP for eight protocols were below ACR AD. The 75th percentile CTDIvol and DLP were below ACR DRL for all protocols. Dose changes were linked to documented protocol changes. C-spine WO and Neck W initially had low CTDIvol and DLP, which were slightly increased to improve image quality. In early 2020, high CTDIvol for Chest PE prompted protocol changes, leading to dose reduction. Kidney WO had DLP slightly increased in 2020 to early 2023, prompted protocol changes and significant dose reduction between Q2 and Q3 2023. For Head WO, median CTDIvol was 9–11% above ACR AD (49 mGy), justified by image quality concerns, while 75th percentile remained below ACR DRL (57 mGy).
Conclusion:
Through rigorous protocol review, the majority of median and 75th percentile dose metrics were lower than ACR AD and DRL, respectively, as of Q4 2024.