“MRI is more accurate than SPECT in detecting the presence of significant coronary artery disease”


Sven Plein, Senior British Heart Foundation Clinical Research Fellow and consultant cardiologist in Leeds, UK, is an expert in cardiovascular magnetic resonance (CMR) imaging. In this interview with Cardiovascular News, Plein spoke about the current trends in CMR. He also spoke about the benefits of this modality and the recent use of CMR as an endpoint in acute myocardial infarction trials.

What are the current trends in cardiac MR for the assessment of ischaemic heart disease and heart failure?


The main current trend in CMR in general is a shift from technical development to robustness and larger trial evidence. In the past, the field focused mostly on incremental improvements of the technology. While these developments are ongoing and important, we are now seeing a period of consolidation and standardisation. For the assessment of ischaemic heart disease in particular we now have a range of accurate, robust and standardised CMR methods such as cine imaging and scar imaging using late gadolinium enhancement. Other methods use different technical solutions, but we now realise that they mostly lead to very similar results.


A good example of this is myocardial perfusion imaging, where different protocols across different platforms provide very similar clinical performance. With perfusion, cine and late gadolinium enhanced CMR, we have a strong set of tools to assess ischaemic heart disease and we are beginning to see how this translates into large clinical trial evidence.


The second major trend I see is the concerted effort of researchers and vendors to make CMR imaging faster and less complicated. All vendors are working on protocols which are almost push-button, with a lot less planning involved and a total scan time of 30 minutes rather than an hour.


A related trend is towards whole heart 3D CMR. Scar imaging can be performed very well in a 3D single breath-hold. Cine imaging can, in principle, be done in a 3D breath-hold although in practice we often still acquire multiple slices because the image quality remains superior.


And recently there have been proposals for 3D myocardial perfusion CMR imaging. I expect that in the next few years we will be seeing really quick, mostly 3D-based protocols with little planning becoming widely available.


One of the benefits of cardiac magnetic resonance is the fact that it works without exposing the patient to ionising radiation. What are the other benefits?


The absence of exposure to harmful radiation is a key benefit of CMR that sets it apart from most other main imaging modalities. Other benefits of CMR depend on the clinical context. In the setting of stable ischaemic heart disease, for example, CMR has several principal advantages. The absence of ionising radiation has already been mentioned. The second advantage is the high-resolution scar imaging with late gadolinium-enhanced methods, which depicts scars with exceptional special definition and anatomical correlation, not matched by any other imaging modality. While it is in principle possible to perform late contrast enhanced imaging with CT, this presently has lower contrast than with CMR and adds radiation exposure to the test.


The third advantage of CMR in stable ischaemic heart disease is that perfusion assessment appears to be more accurate than single-photon emission computed tomography [SPECT].


Our group has recently completed the CE-MARC study, a prospective head-to-head comparison of CMR and SPECT in 750 patients with suspected angina with the main findings due to be published in The Lancet this year. We found that CMR was significantly more accurate than SPECT in detecting the significant coronary artery stenosis on subsequent angiography. These findings are consistent with a previous study, MR-IMPACT, which was published in the European Heart Journal in 2008. There seems to be consistent evidence that CMR has increased diagnostic accuracy compared with SPECT, probably due to the fact that it has much better spatial resolution.


In acute myocardial infarction, the principle of late gadolinium enhancement is very widely used these days to measure myocardial infarction size. Combining this with an assessment of area at risk with T2-weighted MRI allows us to measure area at risk, area infarcted and make predictions of function recovery.


The CRISP-AMI randomised trial, by Patel et al, which was published online in JAMA, in November, used infarct size measured by cardiac MRI as the primary endpoint of the study. Will this type of use of cardiac magnetic resonance become more frequent? What are the benefits?


There is a rapidly growing number of multicentre trials in acute myocardial infarction that use CMR as their main endpoint. CRISP-AMI was one of these studies. The reason for this trend is that CMR is by far the most accurate technique to measure infarct size and contractile function, which means that the sample size required to show differences in endpoints is much lower than with other imaging modalities. In addition, CMR allows an assessment of microvascular damage following acute MI and provides estimates of area at risk. I expect to see many more studies using CMR endpoints similar to CRISP-AMI in the next years.


You have been involved in the study of high spatial resolution myocardial perfusion MRI. How does it work for the detection of coronary artery disease?


As I said earlier, the key developments in CMR at the moment are that we are moving away from small incremental changes in pulse sequences to robust, cross-platform protocols. However, at the same time we continue to push the boundaries of resolution and image speed. High resolution perfusion CMR imaging, meaning acquisition at a resolution in the 1mm range, is an exciting area of development and may well have a future clinical role in particular in detecting balanced or peri-infarct ischaemia. However, to date there is no published comparison between standard resolution and high-resolution perfusion MR, so its role still needs to be defined.


Your group developed a comprehensive cardiac MR protocol for assessment of coronary artery disease. What are the parameters and how does it work?


The comprehensive CMR protocol we published some years ago was used in the CE-MARC study. It included cine imaging, rest and stress myocardial perfusion imaging, late gadolinium enhancement and coronary MR angiography.


Interestingly, in CE-MARC coronary MR did not improve diagnostic accuracy over the other three components, so that we would now not recommend that coronary MR angiography is included in a routine protocol for assessment of stable IHD. This is consistent with the recommendations for standardised protocols published by the Society for Cardiac Magnetic Resonance in the Journal of Cardiovascular Magnetic Resonance in 2008. These suggest for the assessment of ischaemic heart disease a combination of cine imaging for assessment of left ventricular function, rest/stress perfusion for detection of ischaemia and late gadolinium enhancement for detection of scar. In most centres, this would be the standard protocol used today.


In acute myocardial infarction, we would conventionally perform a comprehensive study including T2-weighted imaging, cine imaging and early and late gadolinium enhanced imaging.


How do you see echocardiography versus cardiac MRI? Are they complementary tests?


Yes, absolutely. All the modalities are here to stay – echocardiography, CT, MR and nuclear imaging. What is less clear is which of these modalities is most appropriate to investigate which patients in which clinical scenarios in terms of diagnostic accuracy, but even more importantly in terms of patient outcome and, sadly, cost-effectiveness. It is clear that CMR will not replace echocardiography as the first-line investigation for the majority of the patients with heart disease. But, with stress echocardiography, it offers us an excellent alternative to SPECT and CMR in patients with suspected ischaemic heart disease. However, some patients for example do not have good echocardiographic windows so that an echocardiogram is not the right rest for them. On the other hand, some patients are claustrophobic or have non-compatible metallic implants so an MRI scan is not the right test for them. It is therefore important to have a range of tests available. Both MRI and echo are safe, with no ionising radiation, so that both tests have a very clear future. The more contentious debate is whether CMR could be a safer and potentially more accurate alternative to SPECT imaging. There are indications that this is the case, but more evidence is needed.  


What two questions would you like to see answered in the imaging field in the near future?


My first question would be “What is the role of CMR in guiding treatment and predicting outcome in ischaemic heart disease?” It is no longer considered sufficient for a test to have better diagnostic performance than another test – it also has to improve patient management and patient outcome. For CMR, this means that we need to find out whether its improved image quality and diagnostic accuracy translate into improved patient management and better outcomes. Relevant studies are underway, for example, the MR-INFORM study, which compares angiography-guided with CMR-guided management in a randomised non-inferiority design.


My second question would be on acute myocardial infarction: How can we use the ability of CMR to characterise infarcts to improve patient outcome and patient management? This is an equally intriguing area, in which CMR provides unique information and is likely to make a real impact on patient outcome.