Dr. Frank Rademakers,
A major challenge to the cardiac imaging community is the increased impact of imaging on medical budgets worldwide, without randomised controlled trials to back up the use of imaging in most of the areas where it is being applied for the moment. Such evidence is lacking, not so much with respect to sensitivity and specificity for the condition it is tracking, but definitely with respect to outcome data for individual patients. Prognostic data are less available than sensitivity/specificity data, but more so than outcome data. The different speakers of the session addressed various aspects of this problem. Dr. R. Gibbons, of the Mayo Clinic, Rochester, US, spoke about the tremendous growth of cardiac imaging (slide 1) with a concomitant reduction in reimbursement per act. He suggested 4 actions to be taken:
We thus need to regulate ourselves by managing the growth of cardiac imaging.
Dr. P. Kaufmann from Zurich, Switzerland, illustrated the integration of imaging modalities in clinical practice. This can be accomplished either by sequential imaging and side-by-side analysis or by combined or hybrid imaging; he emphasized that in either case, all clinical data should also be incorporated in the process. Presently, most hybrid imaging concerns nuclear (PET/SPECT) + CT (slide 3), either by software fusion or intrinsic hybrid acquisition of images. Most often the goal is combining anatomy and function; this is warranted since coronary obstructive disease is much more prevalent than ischemia by a factor of 10; while the first alone requires preventive measures, the latter requires tailored treatment. A combination of anatomy and function might provide better prognostic information. He showed examples were hybrid imaging could improve interpretation of the link between stenotic lesions and ischemic regions in the setting of extensive multi-vessel disease (slide 4). Since the relation between degree of stenosis and ischemia is curvilinear and only reaches a maximum of 50% ischemia for 100% stenoses (due to scar and collaterals) (slide 5), it is relevant to quantify this ischemia to tailor therapy.
Dr. R. Martin from Atlanta, US, tackled the problem of evidence about imaging influencing clinical outcomes. While this can be argued for the individual patient on the basis of extensive personal experience, this remains grossly anecdotal, and very limited scientific data is available. Although imaging has evolved tremendously with respect to technical possibilities, indications and usage in clinical practice, the scientific evidence for this usage is not there. Added to this is the problem of litigation, mostly in the US, which induces a defensive over-use of imaging. Similar problems exist in cardiology, oncology, neurology, etc. In this respect, as emphasized by other speakers, the benefit (and cost) to the patient is more important than the accuracy of the test. Overall the lack of good data is most evident for asymptomatic patients; in this respect the use of stress testing could be relevant, also in conditions where this was previously counter indicated, like aortic stenosis, and in mitral regurgitation. In the conclusion, yet again, the importance of imaging trials with outcome data was emphasized. Dr. A. Fraser from Cardiff, UK, addressed the establishment of joint clinical diagnostic services, as put forward in the statement from the imaging groups of the ESC in 2006 (slide 6). The goal is to come up with common diagnostic protocols and pathways by all specialties concerned (cardiology, nuclear medicine, radiology) based on patient outcome data. In the absence of those data, he argued for the use of ischemia testing in view of the very poor relationship of anatomical diameter stenosis with fractional flow reserve and the presence of ischemia. In this respect he pleaded for the use of quantitative functional parameters rather than description of images. Also radiation needs to be taken into account when choosing a specific technique for a specific parameter, like ischemia. To improve the selection of patients requiring ischemia imaging, he strongly promoted the use of pre-test probabilities to guide the process: patient with low pre-test probability (<10-15%) should be left alone and those with high pre-test probability (>85-90%) should proceed to therapy without imaging, leaving the intermediate risk patients for additional testing (slide 7). Furthermore when such a test is of adequate quality, a negative result should be trusted and not followed through with further testing. Finally the type of population should be taken into account when interpreting test results, since this interpretation can vary with impact on patient management. In the final discussion, the need for large RCTs was emphasized but also the problem of getting the funding for such trials, which typically comes from industry in the setting of pharmacological treatments where this evidence is required for registration of a drug. Such a process does not exist to the same extent for technical imaging devices but is being put in place in several countries. This might pave the way to collaboration between the industry involved and the regulatory agencies requesting such information about outcome data. In the mean time, registries could already provide valuable information. It is up to the cardiology community to start acquiring the necessary data.
Cardiac imaging: 2009 challenges
Our mission: To reduce the burden of cardiovascular disease
© 2017 European Society of Cardiology. All rights reserved