Comparison of patient effective doses from multiple CT examinations based on different calculation methods

Area of Science:

• Medical Physics
• Radiology
• Radiation Dosimetry

Background:

• Accurate estimation of effective dose (E) is crucial for managing patient radiation exposure, especially in recurrent computed tomography (CT) examinations.
• Various methods exist for calculating E, but their consistency and reliability for individual patient assessments remain a concern.

Purpose of the Study:

• To compare the effective dose (E) estimations derived from seventeen different methods for patients undergoing recurrent CT examinations.
• To evaluate the variability and uncertainties associated with different E calculation approaches, particularly for varying patient sizes.

Main Methods:

• Seventeen distinct methods were employed to determine the effective dose (E) for individual CT scan phases and the total examination.
• Methods included estimations based on published data, calculations using volume computed tomography dose index (CTDIvol) and dose-length product (DLP) with conversion coefficients, and patient-specific software calculations.

Main Results:

• Effective dose (E) estimations for a single CT phase varied by ratios of 1.3–6.8 (small), 1.2–6.5 (normal), and 1.7–18.1 (large) patients depending on the method.
• Cumulative effective dose (CED) ratios per patient varied from 1.4–2.5 (small), 1.7–4.3 (normal), and 2.2–6.3 (large) across different methods.
• Minimum CED ranged from 38 to 200 mSv, and maximum CED ranged from 122 to 538 mSv, demonstrating wide variations.

Conclusions:

• The method used for calculating effective dose (E) significantly impacts individual patient assessments in recurrent CT examinations.
• Substantial uncertainties, with variations up to 18.1 times for individual E and 6 times for CED, necessitate careful consideration in clinical practice.
• While E is useful for population studies, its application to individual patients requires acknowledging and accounting for the inherent methodological uncertainties.