Author + information
- Isabel Castellano, PhD and
- Edward D. Nicol, MD, MBA∗ ()
- ↵∗Department of Cardiovascular CT, Royal Brompton & Harefield NHS Foundation Trust, London, England
We read with interest the excellent report by Einstein et al. (1) and the accompanying pragmatic editorial (2). We support the laudable aim of involving patients in the discussion of radiation exposure as part of the clinical decision process, but we wish to highlight a couple of areas for cardiovascular computed tomography (CT) specifically that require expert consensus to ensure this process is robust.
Dose calculation in cardiovascular CT is a complex field of medical physics, and the lack of clarity on even the fundamentals clearly needs to be addressed. First, the determination of an effective dose for procedures such as cardiovascular CT remains controversial, while the lack of cardiovascular CT dose reference levels and current benchmarking of institutional practice against these makes generalization of dose potentially meaningless.
Several different conversion factors are currently used in published reports when converting dose-length product into an effective radiation dose (in mSv) delivered to a patient. Aside from the fact that these conversion factors were not designed to be applied to individual patients, the variety of quoted figures requires resolution. Those that are routinely cited in published reports include the thoracic conversion factors of 0.017 or 0.014 mSv/mGycm (3), which estimate the effective dose as defined in International Commission on Radiological Protection publication 60, and specific cardiovascular CT conversion factors of approximately 0.028 (4), which estimate the effective dose as defined in International Commission on Radiological Protection publication 103. Pediatric conversion factors are even more contentious. Because cardiovascular CT does not usually involve scanning the whole of the thorax, and it also predominantly involves exposure over highly radiosensitive breast tissue, an effective dose for any given dose-length product is proportionally higher for cardiac versus thoracic CT (Figure 1).
The issue of dose reference levels for cardiovascular CT is also important, and a recent survey of cardiovascular CT centers in the United Kingdom performed by the British Society of Cardiovascular CT revealed a 5-fold variability in median dose-length product for standard CT coronary angiography from 200 mGycm to more than 1,000 mGycm. Assuming all images were diagnostic and scans were performed on appropriate cardiac enabled scanners, this is alarming. However, given the evidence that the majority of UK centers perform low volumes of scans, with only a handful performing more than 1,000 scans each year (5), this may not be surprising. The challenge, however, is how to determine a suitable dose reference level with such variation in both number and dose.
Only when we, as a clinical and scientific community, have rationalized and agreed on these values will discussions with our patients regarding dose be evidence based and meaningful.
- American College of Cardiology Foundation
- Einstein A.J.,
- Berman D.S.,
- Min J.K.,
- et al.
- Wexler L.