Computed tomography (CT) of the heart combines X-ray images with the aid of a computer to generate cross-sectional views of the body. Cardiac CT uses advanced CT technology with or without intravenous contrast dye to visualise cardiac anatomy, coronary circulation, and the great vessels. Cardiac CT is especially useful in evaluating the myocardium, coronary arteries, pulmonary veins, thoracic aorta, pericardium, and cardiac masses, such as thrombus of the left atrial appendage. Cardiac CT has evolved greatly over the last 20 years, and continues to show tremendously successful development. Technical improvements have translated into new applications and enhanced diagnostic accuracy. These new diagnostic opportunities will be beneficial for many individuals with known or even suspected cardiovascular disease.
Alan Katz, director of medical information technology at St Francis Hospital, talks to European Cardiovascular Disease 2006 about the future of cardiac CT.
Q: Describe the technical advances in the use of CT for the diagnosis of coronary artery disease.
A: In recent years, CT angiography (CTA) is by far the biggest advance, the reason being that the machines are now capable of getting better resolutions at faster rates. In about 1984, a machine called the electron beam CT scanner was introduced and that machine used a fixed detector with an electron beam that rotated the X-ray beam. That machine was able to give a good resolution of about 26.6 x (about) 1 sonometer in diameter. However, the z-axis would not give a high enough resolution. The concern with this is that the coronary arteries are the region of 0.75-1mm in diameter. In 1999, the multidetector CT scanner was introduced. Initially, it was a four-slice scanner, which in 2004 became a 16-slice scanner, and recently in 2005 became a 64-slice scanner. These machines differed in that the X-ray source was rotating around the patient, and this X-ray source would move very quickly. In fact, conventional machines now get 64 slices per view and have a rotation speed of about 330 milliseconds, which allows the physician to see very fast-moving objects. It effectively slows down very fast-moving objects and the fact that it can spin that quickly gives resolutions in the order of 0.4x0.4x0.6mm, which allows a clear view of the coronary arteries.
So, in summary, the mulitdetector CT scanner overcame two technical problems. The improved spatial resolution meant the coronary artery could be viewed, and the improved temporal resolution allowed the physician to image a moving object, such as the heart or the coronary arteries, very quickly, allowing stop-frame – something that was particularly useful when viewing the right coronary artery, which moves quickly. It also allowed the patient to only hold their breath for 5-10 seconds.
Q: What do you see as the most important advances in CT technology today?
A: The advances in multi-detector CT technology that offer better spatial and temporal resolution have now allowed cardiologists to see the coronary arteries without having to perform invasive procedures such as coronary artery catheterisation. So that clearly is the biggest advance in the new technology from a cardiologist's point of view.
Q: What about future technologies?
A: Technology is moving very quickly, and manufacturers are working to rectify some of the limitations of the technique. One such limitation is the fact that there is a fair amount X-ray radiation involved in CT scanning. It's around 10-11 millisieverts, which is comparable to a nuclear stress test. So although it's not an inordinate amount, it would be good to reduce the radiation level. Some manufacturers are introducing techniques to lower the level of radiation to the equivalent of an electrocardiogram (ECG), and this has been somewhat helpful. There are other technologies that manufacturers are exploring to try and decrease the radiation dose further.
The second problem manufacturers and software companies are addressing is that of coronary calcification. When calcification is identified in the coronary arteries it indicates atherosclerosis. To date, we don't know of anything else that causes calcium in the coronary arteries, other than atherosclerosis. However, when there is calcium in the coronary arteries, we can see the soft plaque or the extent of stenosis beneath it in an artefact that is called a 'blooming' artefact. In my opinion, there will be help in dealing with the difficulty of looking at calcified coronary arteries in the next several years. However, it isn't on the horizon, and it's unlikely that anything will be introduced for six months that will allow cardiologists to use this technology in a greater population base. Today, patients who have significant coronary artery disease (CAD) often have significant calcification and, although they are at risk of increased cardiac events, we cannot predict how tight the stenosis is underneath the calcified lesions.
Third, the fact that cardiac magnetic resonance imaging (MRI) is still trying to look at coronary arteries gives the possibility of a competing technology. Cardiac MRI has not yet been able to get the spatial resolution necessary to see coronary arteries, but there is continued work being undertaken. At present, it's unclear exactly how that work will progress, but both cardiac MRI and cardiac CT technologies are looking towards ways of carrying out better plaque characterisation. For instance, today with CTA we identify either calcified plaque or soft plaque.
However, there is more to a plaque than whether or not it contains calcium. The important questions to ask are how much fibrosis is there? And how unstable is that plaque? Both of these technologies are going to endeavour to see whether they can improve plaque characterisation and try to predict which plaques are most likely to rupture and cause a cardiac event, and that will be exciting to watch.
Q: How does the use of CT compare with traditional X-ray angiography?
A: Traditional X-ray angiography involves an invasive cardiac catheterisation procedure that entails actually putting a tube into the heart and directly cannulating the coronary arteries. Usually this procedure is performed from the groin going up the femoral artery, although in some laboratories it's done from the arm going through the radial artery. In any case, it involves bringing a patient in, putting catheters into the arteries and actually moving these catheters up into the coronary artery. So it's a much more invasive procedure than CTA, and with this invasion there are added complications. As a cardiologist, during an X-ray angiography procedure you always worry about hurting a coronary artery and you worry about hurting the leg if you're catheterising from the leg. You need to have more people in the room watching the patient, so it's a much more time- and resource-consuming procedure. However, despite these drawbacks, the spatial and temporal resolution of invasive angiography is still currently superior to CTA. It also shows a moving view of the coronary vessel, again because the temporal resolution is higher. In addition there is the opportunity to perform therapeutic intervention, such as balloon angioplasty and stent placement, during the catheterisation procedure.
Cardiologists will be looking at the subgroups of patients who will benefit from CTA over invasive angiography. What is clear is that patients who have a low likelihood of CAD and fit certain criteria (the ability to hold their breath for 5-10 seconds, to get their heart rate down to around 60bpm and have a regular heart beat), who go to the CTA lab and have a normal study, are very unlikely to have CAD. So these patients can avoid catheterisation and, in the future, it will be clear where these two techniques fit in.
To contrast CTA with conventional stress testing, typically today a cardiologist would perform a stress test on a patient (either an ECG, echocardiogram (Echo) or nuclear test). To use the example of SPECTIVE, which is a nuclear technique, the sensitivity is about 87% and the specificity is about 73%. So the challenge is firstly to pick up those patients who had CAD who were missed by the nuclear test, who really would benefit from the cardiac catheterisation procedure, and to keep that 27% of patients who have normal coronary arteries out of the cath lab. Not only are these people subjected to the threat of increased complications; it actually costs the system an estimated US$4 billion (in the US alone).
When considering CTA, the negative predictive value is above 95%, whether you evaluate by patient, by artery or by segment. The positive predictive value varies depending on the study, but it's not quite as good. The positive predictive value of the 64-slice is between 85% and 90% and, again, it depends on how you look at it. Articles are just being published with data on the 64-slice CT and, again, the values may vary more depending on whether you're talking about segment, artery or patient. One of the things that does need to be undertaken is improvment of the positive predictive value, the reason being that we have a lot of difficulty seeing all the segments, particularly in the presence of calcification. CTA can also help identify anomalous coronary arteries; their identification is needed as some of these patients are at risk of sudden death and obviously need to know about it. Also, when in the cardiac cath lab, at times the operator is unable to find where a coronary artery is and CTA can actually show where the artery is and identify whether it is clean or whether there's a problem with it. The other situation where the use of CTA is useful is during a bypass graft procedure. It is sometimes very difficult and time-consuming to cannulate bypass grafts and if there are any complications after this surgery, you do not want to bring a patient back to the cardiac cath lab. CTA shows the bypass graft very well, and is a great alternative.
Q: What is your vision of where the trends are heading for cardiologists?
A: Clearly the trend is going to be improving prediction and prevention of CAD and valvular heart disease. In general, cardiologists are striving to find better ways of predicting who is at risk of a cardiac event, and offering early treatment in a cost-effective manner. We are looking towards a future where studies will be carried out to compare different modalities so that patients are offered the correct and most cost effective technologies. It is impossible to perform every imaging study on every patient as the cost soon mounts. We need to be able to ascertain which patients benefit from which imaging studies and which patients benefit from which blood studies. There is also a great deal of work being carried out looking at markers of CAD such as fibrinogen. This work on genetic markers is continuing a current trend to predict CAD and cardiac disease in general a lot earlier and be able to apply therapy to those patients before they progress to the point of having had myocardial infarction or heart failure.