CardioExchange Archive

The European Society of Cardiology (ESC) recently released a position paper on radiation safety in cardiovascular imaging. Here is the paper’s stated objective:

[to] provide a European perspective on the best way to play an active role in implementing in clinical practice the key principle of radiation protection that: ‘each patient should get the right imaging exam, at the right time, with the right radiation dose’.

Clearly written and highly readable, the paper touches broadly on key issues related to radiation safety, including the principles of justification (ensuring that imaging studies are clinically necessary and appropriate) and optimization (ensuring that imaging is performed in the safest manner). However, the paper is strikingly lacking in specific and implementable recommendations and, as such, seems to fall short of its stated objective. It clearly communicates the need for improving radiation safety in cardiovascular imaging but does little to help us understand how we can do that.

For example, the document advocates for justification — a key principle of radiation safety — but does not set forth an action plan for promoting it. How do we monitor appropriateness of imaging studies? How do we reduce the number of inappropriate studies? What are the barriers to using tools such as the Appropriate Use Criteria (AUC) for this purpose, and how can those barriers be overcome? While there is scant evidence in the literature to help answer these difficult questions, an attempt to address these issues would have been useful, particularly given the breadth of expertise among the document authors.

Similarly, the document briefly mentions a few optimization techniques for nuclear cardiology and cardiac CT but does not discuss practical strategies to promote their widespread adoption by imaging labs. How should performance of imaging labs be monitored? How do we address the lack of diagnostic reference levels for most types of cardiac imaging with ionizing radiation?

As another example, the paper highlights the knowledge gap among clinicians regarding typical doses for cardiac imaging studies and the risk attributable to those doses. It is suggested that requiring discussion of radiation-related risks as part of informed consent will force clinicians to bridge that knowledge gap. Although there is truth in this statement, the strategy does not seem to be adequate for addressing the problem systematically.

Some inaccurate statements in the paper are also troubling. For example, it states, “at least one-third of all cardiac examinations are partially or totally inappropriate.” First, the percentages of inappropriate studies in the cited study were actually 14% and 18% for stress echo and SPECT, respectively. Characterizing the uncertain category in AUC as “partially inappropriate” is incorrect and not consistent with its definition. Furthermore, it is unclear how the authors’ inference about all cardiac imaging is justified on the basis of data from a relatively small study of two imaging modalities from a single institution.

More important, some of the cancer-risk estimates listed in the paper are controversial, including those quoted for interventional cardiologists and children with congenital heart disease. The cited studies do not use the preferred method of cancer-risk estimation, which involves calculation of organ doses, applying tissue weighting factors, and then applying age- and sex-specific risk coefficients from the BEIR VII report. Rather, those studies derived risk estimates based on effective doses, which is beyond the intended scope of the effective-dose concept and is not scientifically valid. Effective dose should not be used for epidemiologic studies or to estimate population risks, given the inherent uncertainties and oversimplifications (Martin et al., Br J Radiol 2007; 80:639; McCullough et al., AJR 2010; 194:890).

Finally, the paper makes no substantial mention of the European experience. Discussing success stories from Europe and approaches that European centers have taken to improve radiation safety would have been very helpful.

Overall, considering the large audience that any ESC document will reach, I believe this paper will significantly raise general awareness of radiation safety among cardiologists and other clinicians. In some ways, though, it is also a missed opportunity to influence clinical practice more meaningfully.

If you have a chance to read the document, please share your thoughts: Does it give you new insights about radiation safety? Will it influence your clinical decision-making? If so, how?