In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

Statement From The ESC Regarding The Ongoing CT-scan vs. Invasive Angiography Debate

Noninvasive imaging such as CT angiography is complementary to invasive angiography – they do not directly compete with each other

Recent media reports based on a new study published in the New England Journal of Medicine (27 November 2008 issue) have announced that “noninvasive CT scans cannot replace imaging of coronary artery blockages by conventional invasive angiography”.

In response, imaging specialist Professor Juhani Knuuti, spokesperson for the European Society of Cardiology and Past-Chair of the ESC Working Group on Nuclear Cardiology and Cardiac CT, emphasizes the following points.

The CT imaging techniques have developed very rapidly and during recent years the use of CT to image coronary arteries has been rapidly growing. While it was initially hoped that this noninvasive test could even replace invasive coronary angiography, it has been obvious for couple of years that this is not the case.

The strength of invasive coronary angiography is that the treatment procedures such as dilation of stenosis and stenting can be done in the same session, which is never the case with noninvasive imaging studies. However, due to risks of invasive procedures, these tests should be ideally limited to the patients with high likelihood of disease and necessary procedures.

The strength of CT angiography in the previous studies was the very high negative predictive value i.e. when the result of test is negative the significant coronary artery disease is very unlikely. Consequently, ESC imaging experts have recommended CT angiography to be at the best in excluding coronary artery disease in the patients with low to moderate pre-test likelihood of disease (Eur Heart J. 2008; 29(4):531-56).

The debated study by JM Miller and colleagues (N Engl J Med. 2008; 359(22):2324-2336) is the first multicentre trial for 64-slice CT angiography. While the results were mainly in agreement with the previous single centre studies and demonstrated good accuracy in general, the negative predictive value was somewhat lower (83%). This raises questions whether the technique is indeed very precise in excluding coronary artery stenoses when applied in clinical routine and in multiple sites.

Another issue that was actively discussed, was the radiation risks of the CT scans. It is obvious that all necessary actions are needed to minimize the risks of imaging, including radiation risks. In the debated study, the radiation doses by CT were 2-3 fold higher than in invasive angiography, but still in moderate level. However, with the current generation of CT scanners, which were not widely available at the time the study was started, the radiation dose is ranging from 3 to 7 mSv, which is actually less than with invasive angiography. Newer techniques are looking at lowering the radiation dose even further.


While waiting further studies and confirmation of these results, it is important to emphasize one completely ignored issue in the current debate. The fact is that anatomical imaging, noninvasive or invasive, does not accurately tell whether the detected coronary blockages are causing ischemia and whether the lesions are the cause of the symptoms of the patients. As it is stated clearly also by clinical guidelines, functional ischemia testing is needed to assess whether the coronary stenoses are hemodynamically significant i.e. how much blockage prevents the blood from flowing. This can be done by cardiac nuclear imaging, stress echocardiography, magnetic resonance imaging or invasive flow measurements. Using CT, now also noninvasive imaging of coronary anatomy is possible and it is very likely that this information will be combined with functional imaging findings making finally comprehensive noninvasive evaluation possible.